Anda di halaman 1dari 117

RINGKASAN MATERI KULIAH

SEMINAR AUDIT
“THE TROLLEY DODGERS”
Dosen Pengampu: Dr. Maria Mediatrix Ratna Sari, S.E., M.Si., Ak., CA

Oleh:

Kelompok 3

Ni Ketut Ayu Rosiana Dewi (1981621010)


Prema Sanjaya (1981621023)
Ni Luh Putu Ayu Lastri Pramiswari (1981621024)
I Wayan Megayana (1981621025)

PROGRAM STUDI MAGISTERAKUNTANSI


FAKULTAS EKONOMI DAN BISNIS
UNIVERSITAS UDAYANA
DENPASAR
2021
RINGKASAN KASUS

Pada tahun 1890, tim bisbol profesional Brooklyn Dodgers Trolley bergabung
dengan Liga Nasional. Setelah kesulitan bersaing dengan tim sekota mereka, New York
Yankees, selama 7 dekade lamanya, mereka memutuskan untuk pindah ke Los Angeles
pada tahun 1958. Sang pemilik, Walter O’Malley, melihat kesempatan membuat timnya
lebih berkembang dan menguntungkan dengan pindah ke Los Angeles.
Sejak pindah ke Los Angeles, Dodgers telah membuat iri dunia bisbos. Mereka
menjadi sebuah prototipe bagaimana waralaba harus dijalankan. Selama tahun 1980,
Dodgers menjadi franchise paling menguntungkan di dunia bisbol dengan marjin
keuntungan sebelum pajak mendekati 25 persen selama bertahun-tahun.
Pada akhir tahun 1997, Peter O'Malley, putra dan pemilik utama Dodgers menjual
franchise sebesar $350 juta kepada Rupert Murdoch. Juru bicara Murdoch memuji
keluarga O'Malley untuk dalam menetapkan standar emas untuk kepemilikan waralaba.
Dalam sebuah wawancara sebelum Peter O’Malley menjual Dodgers, keberhasilan
organisasinya disebabkan oleh para ahli yang telah dipertahankan dalam seluruh bidang
fungsional. Edward Campos, seorang akuntan yang lama berkerja untuk Dodgers, adalah
contoh sempurna dari salah satu ahli dalam organisasi Dodgers. Campos memulai karir
di Dodgers sebagai seorang pemuda. Pada 1986, setelah hampir dua dekade dengan klub,
ia telah meniti karier hierarki untuk menjadi seorang payroll chief.
Setelah mengambil alih payroll department Dodgers, Campos merancang dan
menerapkan sistem penggajian yang baru. Sistem yang kabarnya hanya dia yang
sepenuhnya mengerti. Bahkan, Campos mengendalikan sistem ini sendiri, sehingga dia
secara pribadi dapat mengisi kartu gaji mingguan untuk 400 karyawan Dodgers. Campos
dikenal tidak hanya untuk etos kerjanya, tetapi juga karena kesetiaannya kepada klub dan
sang pemilik. Dodgers sangat percaya padanya. Bahkan ketika ia sedang berlibur, ia
kembali hanya untuk melakukan daftar gaji.
Sayangnya, kepercayaan Dodgers disalahgunakan oleh Campos. Selama periode
beberapa tahun, Campos menggelapkan ratusan ribu dolar dari sang pemilik. Menurut
catatan pengadilan, Campos menambahkan karyawan fiktif untuk berbagai departemen.
Selain itu, Campos rutin menggelembungkan jumlah jam kerja beberapa karyawan dan
kemudian membagi penghasilan tersebut fifty-fifty dengan orang yang bersangkutan.

1
Skema penipuan ini terungkap ketika usus buntu menimpa Campos, sehingga
memaksa controller Dodgers untuk mengambil alih tangung jawabnya sementara. Saat
menyelesaikan daftar gaji satu minggu, controller melihat bahwa beberapa karyawan,
termasuk penerima tamu , penjaga keamanan, dan tenaga penjualan tiket, dibayar dalam
jumlah yang luar biasa besar. Dalam beberapa kasus, karyawan penghasilan $7 per jam
menerima gaji mingguan mendekati $2.000. Setelah dilakukan investigasi kriminal dan
pengajuan tuduhan terhadap Campos dan para pengikutnya, semua individu yang terlibat
dalam kasus penipuan penggajian ini mengaku. Sebuah pengadilan negara menghukum
Campos delapan tahun penjara dan mengharuskan dia untuk membayar ganti rugi sekitar
$132.000 keada Dodgers. Konspirator lainnya juga menerima hukuman penjara.
Individu-individu tersisa yang terlibat dalam skema penggajian membuat restitusi, dan
ditempatkan dalam masa percobaan selama di penjara.

PERTANYAAN KASUS
1. Identifikasi tujuan audit utama untuk fungsi payroll klien. Mengomentari kedua
tujuan terkait dengan tes kontrol dan yang terkait dengan prosedur audit substantif.
Jawaban:
Tujuan audit terhadap fungsi payroll klien secara keseluruhan adalah untuk
mengevaluasi apakah saldo akun yang dipengaruhi oleh adanya fungsi ini disajikan
secara wajar sesuai dengan prinsip akuntansi yang berlaku umum atau Generally
Accepted Accounting Principle (GAAP). Namun, khusus untuk audit terhadap fungsi
payroll difokuskan terhadap tujuan audit yang berkaitan dengan transaksi
(transaction-related audit objectives). Hal ini dikarenakan:
a. Dalam fungsi payroll hanya terdapat satu kelas transaksi, karena penerimaan jasa
yang diberikan oleh pegawai dan pembayaran atas jasa tersebut melalui payroll
biasanya terjadi di dalam suatu periode yang pendek.
b. Transaksi ini tidak begitu signifikan dibandingkan dengan saldo di dalam neraca
atau dibandingkan dengan jumlah keseluruhan dari transaksi-transaksi yang
terjadi di dalam satu tahun.

2
2. Apa kelemahan kontrol internal yang terbukti dalam sistem payroll Dodgers?
Jawaban:
Kontrol internal adalah suatu proses, yang dipengaruhi oleh dewan direksi,
manajemen, dan pegawai lainnya dalam suatu entitas, yang dirancang untuk
memberikan keyakinan yang memadai (reasonable assurance) tentang pencapaian
tujuan dalam keandalan pelaporan keuangan, efektivitas, dan efisiensi operasi dan
kepatuhan terhadap hukum dan peraturan. Kelemahan kontrol internal dalam sistem
payroll Trolley Dodgers berasal dari :
a. Desain Kontrol Internal
Desain kontrol internal perusahaan dilakukan oleh Campos, kepala operasi
penggajian, dimana Campos telah merancang pengendalian internal oleh dirinya
sendiri dan menerapkan sistem payroll baru yang hanya dapat dipahami
sepenuhnya olehnya. Seperti yang kita ketahui tujuan dari desain kontrol internal
adalah untuk mencegah dan mendeteksi adanya salah saji material dalam laporan
keuangan. Berdasarkan kasus ini resiko yang terjadi adalah Campos dapat dengan
mudah melakukan penipuan atau fraud karena tidak ada orang lain yang bisa
memahami desain pengendalian internal kecuali Campos sendiri.
b. Tidak Adanya Kontrol yang Independen terhadap Campos
Campos mengontrol sistem yang dibuatnya sehingga Campos secara personal
dapat mengisi kartu gaji mingguan dari 400 pegawai perusahaan. Campos dapat
mencatat setiap pernyataan palsu karena tidak ada bukti yang menunjukkan
apakah catatannya dapat dipercaya atau tidak.
c. Tidak Adanya Pemisahan Tugas (Separation of Duties)
Campos mengambil keuntungan dari otoritas yang dimiliki karena dia adalah
pekerja yang diberikan terlalu banyak kepercayaan oleh perusahaan. Ketika
Campos sedang berlibur, dia kembali dan mengerjakan tugas seluruh tugas
payrollnya, hal ini mengindikasikan tidak adanya pemisahan penyimpanan aset
dari proses akuntansi. Seseorang yang secara temporer atau permanen menyimpan
aset seharusnya tidak diperkenankan untuk menghitung aset tersebut. Hal ini
dikarenakan dengan memperbolehkan satu orang untuk mengerjakan kedua fungsi
tersebut akan meningkatkan resiko orang tersebut mengambil aset perusahaan
untuk keuntungan sendiri dan menyesuaikan catatan aset tersebut untuk menutupi

3
tindakan kecurangan ini. Seperti yang kita lihat di dalam kasus Campos adalah
satu-satunya orang yang bertanggung jawab untuk semua tugas yang berkaitan
dengan payroll.
d. Kelemahan Lingkungan Kerja
Di dalam siklus payroll perusahaan terjadi kolusi antara pekerja dengan
manajemennya. Hal ini menunjukkan bahwa Campos dan pengikutnya memiliki
etos kerja yang rendah. Campos secara rutin menggelembungkan jumlah jam
kerja dari beberapa karyawan dan kemudian membagi hasil kelebihan
pembayarnnya 50:50 dengan orang-orang tersebut. Selain itu, Campos juga
menggelapkan beberapa ratus ribu dollar dari majikannya.

3. Identifikasi prosedur audit yang mungkin mengarah pada penemuan skema penipuan
yang didalangi oleh Campos.
Jawaban:
Prosedur audit adalah instruksi yang detail untuk pengumpulan jenis bukti audit yang
akan diperoleh selama melakukan proses audit.
a. Auditor Harus Memahami Flow Chart dengan Jelas
Auditor harus memahami alur flow chart dari sistem payroll Dodgers dengan
jelas. Dari pemahaman itu auditor dapat melacak bagaimana sistem payroll
berjalan di lapangan. Kemudian auditor akan melihat tidak adanya pemeriksaan
yang independen sebagaimana yang telah dilakukan oleh Campos. Departemen
payroll juga tidak memiliki pemisahaan penugasan atau segregation of duties dan
semua pekerjaan hanya dilakukan oleh Campos seorang.
b. Auditor Dapat Mewawancarai Karyawan di Departemen Payroll
Auditor dapat mewawancarai karyawan di departemen payroll dengan
mengajukan pertanyaan apapun yang mereka pikir sangat penting dan perlu untuk
diketahui. Misalnya, berapa banyak Campos membayar operasi sehari-hari untuk
setiap tingkat staff dan bagaimana cara Campos mengelola time card setiap
pegawai.
c. Tes Auditor Bagi Karyawan yang Fiktif
Untuk memastikan semua transaksi yang telah dicatat oleh Campos dan
departemen payroll benar-benar ada dan mereka melakukan pembayaran kepada

4
karyawan yang no-fiktif, auditor dapat membandingkan nama di dalam cek yang
dibatalkan dengan time card yang telah dibuat. Auditor dapat memindai bukti-
bukti pendukung dalam cek yang dibatalkan. Selain itu, Auditor juga dapat
membuat perbandingan file pribadi karyawan dengan time card yang sudah
dibuat. Selain itu, auditor dapat menjalankan uji kontrol dengan transaksi yang
dipilih di departemen payroll dan dengan departemen SDM. Auditor dapat
memilih transaksi dalam seminggu dan memeriksa cek 50 pertama dengan jumlah
yang paling besar, kemudian pilih cek secara acak atau memilih cek dimana
auditor berfikir bahwa eror atau kesalahan kemungkinan besar dapat terjadi.
Selain itu, auditor dapat meminta pembayaran payoff secara tiba-tiba kepada
departemen payroll. Ketika auditor meminta hasil pembayaran payroll secara
tiba-tiba maka dapat dipastikan departemen payroll tidak memiliki banyak
kesempatan dan waktu untuk membuat koreksi di dalam sistem mereka. Oleh
karenanya, auditor dapat melihat transaksi yang benar-benar terjadi yang dibuat
oleh departemen tersebut. Akhirnya, auditor dapat melacak adanya salah saji,
penipuan, dan kesalahan dengan benar.
d. Tes Auditor Terhadap Jam Kerja yang Fiktif
Auditor juga dapat melakukan rekonsiliasi terhadap total jam kerja yang
dibayrakan di dalam catatan payroll dengan catatan independen atas jam kerja.
Auditor akan melihat apakah pencatatan transaksi penggajian benar-benar
didasarkan pada time card atau tidak. Auditor dapat melacak jika terdapat indikasi
penipuan atau kesalahan yang disengaja pada pencatatan pegawai.
e. Auditor Membandingkan Tingkat Gaji dengan Posisi yang Sama Dalam Tim
Baseball Lain yang Terdapat Dalam Liga
Auditor juga harus memahami sifat bisnis kliennya dan membuat perbandingan
dengan sifat bisnis lainnya yang sejenis. Dari hasil tersebut auditor dapat
mengetahui berapa tingkat upah yang benar yang biasanya dibayarkan kepada
profesional baseball. Setelah melakukan aktifitas tersebut auditor dapat
mengidentifikasi tingkat upah yang digunakan oleh Campos untuk setiap level
pekerja, apakah relevan atau tidak.

5
f. Auditor Memeriksa Cek Gaji yang Berjumlah Besar
Langkah terakhir, dalam menemukan fraud atau penipuan yang terjadi auditor
dapat memeriksa cek keluar atas pembayaran gaji yang paling besar. Auditor
dapat merekonsiliasi cek antara pencatatan payroll, time card, dan semua data
yang relevan. Auditor akan melihat transaksi mana yang fiktif yang dibuat oleh
Campos.

6
NAMA : NI KETUT AYU ROSIANA DEWI

NIM : 1981621010

REVIEW ARTIKEL INTERNASIONAL

Judul : Faktor-Faktor Yang Mempengaruhi Terjadinya Kecurangan Akuntansi


Dalam Perspektif Fraud Diamond (Studi Empiris pada Bank Perkreditan
Rakyat (BPR) Kabupaten Buleleng)
Penulis : Ketut Budiartini, Gusti Ayu Rencana Sari Dewi, Nyoman Trisna Herawati
Tahun : 2019
Jurnal : Jurnal Ilmiah Mahasiswa Akuntansi Universitas Pendidikan Ganesha,
Vol: 10 No: 2 Tahun 2019 e-ISSN: 2614 – 1930

1. Latar Belakang
Kecurangan (fraud) adalah salah satu penipuan yang sengaja dilakukan
sehingga dapat menimbulkan kerugian tanpa disadari oleh pihak yang dirugikan
tersebut dan memberikan keuntungan bagi pelaku kecurangan. Singelton dan Aaron
J. (2010), mendefinisikan kecurangan (fraud) sebagai perbuatan yang mencakup akal
muslihat, kelicikan, dan tidak jujur dan cara-cara yang tidak layak/wajar untuk
menipu orang lain untuk keuntungan diri sendiri, sehingga menimbulkan kerugian
bagi pihak lain. Dampak yang dapat diakibatkan oleh adanya kecurangan (fraud)
antara lain, hancurnya reputasi organisasi, kerugian organisasi, rusaknya moralitas
karyawan serta dampak-dampak negatif lainnya.
Kecurangan akuntansi telah terjadi di berbagai sektor perbankan. Bahkan di
Bali telah terjadi beberapa kasus terkait kecurangan akuntansi di Bank Perkreditan
Rakyat (BPR). Ketidaksehatan BPR ini disebabkan oleh tingginya tingkat
kecurangan yang dilakukan oleh internal BPR itu sendiri. Salah satu kasus yang
sangat ramai diperbincangkan saat ini adalah kasus penggelapan dana nasabah yang
dilakukan oleh salah satu karyawan pada PD. BPR Bank Buleleng 45 Kas Seririt
yang menelan kerugian senilai lebih dari Rp 635 juta (Singaraja, Balipost.com).
Kemudian kasus penggelapan dana yang terjadi pada tahun 2015 adalah kasus yang
dilakukan oleh Kolektor PT. BPR Suryajaya Kubutambahan, Kecamatan
Kubutambah an, Kabupaten Buleleng, Bali, Jhonieko Maralia Siden alias Joni. Ia
diamankan karena terbukti menggelapkan (menilep) uang nasabah BPR senilai Rp
200 juta (Bali, Tribunnews.com). Penelitian ini berusaha untuk menggali persepsi
para karyawan untuk mengetahui faktor-faktor yang menyebabkan terjadinya
kecurangan (fraud) di sektor perbankan khususnya di Bank Perkreditan Rakyat
(BPR) Kabupaten Buleleng dengan menggunakan konsep fraud diamond.

2. Tujuan Penelitian
Berdasarkan latar belakang tersebut, maka tujuan dari adanya penelitian ini
adalah untuk mengetahui pengaruh dari posisi jabatan, keefektifan pengendalian
internal, penegakan peraturan, kesesuaian kompensasi dan budaya organisasi
terhadap kecurangan akuntansi pada Bank Perkreditan Rakyat (BPR) di Kabupaten
Buleleng.
3. Kajian Teoritis
Kajian teoritis dalam penelitian ini adalah theory frauds diamond, posisi jabatan,
keefektifan pengendalian internal, penegakan peraturan, kesesuaian kompensasi,
budaya organisasi,kecurangan akuntansi
4. Kajian Empiris
Kajian empiris adalah kajian yang diperoleh dari observasi atau percobaan.
 I.G.A. Ngurah Rahma Y. (2017) menyebutkan bahwa posisi jabatan merupakan
hak yang formal dan sah untuk mengeluarkan wewenang dalam mengambil
keputusan, mengeluaran perintah, dan mengalokasikan sumber daya agar tercapai.
Terkadang posisi jabatan yang dimiliki digunakan secara tidak bijak untuk
memperoleh keuntungan individu. Jadi semakin tinggi posisi jabatan yang
diperoleh maka semakin besar pula tindakan kecurangan (fraud) yang dapat
terjadi.
5. Hipotesis Penelitian :
H1: : Posisi jabatan berpengaruh signifikan positif terhadap kecurangan akuntansi.
H2: Keefektifan Pengendalian Internal berpengaruh signifikan negatif terhadap
kecurangan akuntansi.
H3: Penegakan Peraturan berpengaruh signifikan negatif terhadap kecurangan
akuntansi
H4 : kesesuaian Kompensasi berpengaruh signifikan negatif terhadap kecurangan
akuntansi
H5: Budaya Organisasi berpengaruh signifikan negatif terhadap kecurangan
akuntansi
6. Metode Penelitian
Metode yang digunakan dalam penelitian ini adalah metode kualitatif
a. Metode Pengumpulan Data
Pengumpulan data menggunakan non probabability sampling. Data yang
digunakan dalam penelitian ini adalah data primer berupa kuisioner yang
disebarkan secara langsung ke 17 BPR di Kabupaten Buleleng.
b. Sampel
Sampel yang digunakan dalam penelitian ini sebanyak 95 responden.
c. Teknik Analisis Data
Analisis data dalam penelitian ini menggunakan model analisis regresi linier
berganda.
7. Hasil Penelitian
a. Pengaruh Posisi Jabatan terhadap Kecurangan Akuntansi
Hasil pengujian hipotesis pertama mengenaipengaruh posisi jabatan terhadap
kecurangan akuntansi menunjukkan nilai koefisien regresi 0,124 dengan nilai
signifikansi uji t sebesar 0,037. Oleh karena itu, hipotesis pertama dalam
penelitian ini diterima. Hal ini menunjukkan bahwa posisi jabatan berpengaruh
positifsignifikan terhadap kecurangan akuntansi. Jika posisi jabatan semakin
tinggi, maka kecurangan akuntansi juga semakin tinggi.
b. Pengaruh Keefektifan Pengendalian Internal terhadap Kecurangan
Akuntansi
Hasil pengujian hipotesis kedua mengenai pengaruh keefektifan pengendalian
internal terhadap kecurangan akuntansi menunjukkan nilai koefisien regresi -
0,130 dengan nilai signifikansi uji t sebesar 0,026. Oleh karena itu, hipotesis
kedua dalam penelitian ini diterima. Hal ini menunjukkan bahwa keefektifan
pengendalian internal berpengaruh negatifsignifikan terhadap kecurangan
akuntansi. Jika keefektifan pengendalian internal semakin baik, maka kecurangan
akuntansi semakin rendah.
c. Pengaruh Kesesuaian Kompensasi terhadap Kecurangan Akuntansi
Hasil pengujian hipotesis keempat mengenai pengaruh kesesuaian kompensasi
terhadap kecurangan akuntansi menunjukkan nilai koefisien regresi -0,161
dengan nilai signifikansi uji t sebesar 0,042. Oleh karena itu, hipotesis keempat
dalam penelitian ini diterima. Hal ini menunjukkan bahwa kesesuaian kompensasi
berpengaruh negatifsignifikan terhadap kecurangan akuntansi. Jika kesesuaian
kompensasi semakin baik, maka kecurangan akuntansi semakin rendah.
d. Pengaruh Penegakan Peraturan terhadap Kecurangan akuntansi
Hasil pengujian hipotesis mengenai pengaruh penegakan peraturan terhadap
kecurangan akuntansi menunjukkan nilai koefisien regresi -0,140 dengan nilai
signifikansi uji t sebesar 0,035. Oleh karena itu, hipotesis ketiga dalam penelitian
ini diterima. Hal ini menunjukkan bahwa penegakan peraturan berpengaruh
negatifsignifikan terhadap kecurangan akuntansi. Jika penegakan peraturan
semakin baik, maka kecurangan akuntansi semakin rendah.
e. Pengaruh Budaya Organisasi terhadap Kecurangan Akuntansi
Hasil pengujian hipotesis kelima mengenai pengaruh budaya organisasi terhadap
kecurangan akuntansi menunjukkan nilai koefisien regresi -0,129 dengan nilai
signifikansi uji t sebesar 0,033. Oleh karena itu, hipotesis kelima dalam penelitian
ini diterima. Hal ini menunjukkan bahwa budaya organisasi berpengaruh
negatifsignifikan terhadap kecurangan akuntansi. Jika budaya organisasi semakin
baik, maka kecurangan akuntansi semakin rendah.
8. Kesimpulan Penelitian
Berdasarkan data yang telah dikumpulkan dan hasil pengujian yang telah dilakukan
terhadap permasalahan, maka dapat diambil kesimpulan sebagai berikut:
1. Posisi jabatan yang diperoleh oleh seorang karyawan berpengaruh signifikan
positif terhadap kecurangan akuntansi pada Bank Perkreditan Rakyat di
Kabupaten Buleleng dengan nilai koefisien regresi 0,124 dengan nilai
signifikansi uji t sebesar 0,037.
2. Keefektifan pengendalian internal yang diterapkan pada Bank Perkreditan
Rakyat di Kabupaten Buleleng berpengaruh signifikan negatif terhadap
kecurangan akuntansi pada Bank Perkreditan Rakyat di Kabupaten Buleleng
dengan nilai koefisien regresi -0,130 dengan nilai signifikansi uji t sebesar 0,026.
3. Penegakan peraturan yang ada pada Bank Perkreditan Rakyat di Kabupaten
Buleleng berpengaruh signifikan negatif terhadap kecurangan akuntansi pada
Bank Perkreditan Rakyat di Kabupaten Buleleng dengan nilai koefisien regresi -
0,140 dengan nilai signifikansi uji t sebesar 0,035.
4. Kesesuaian kompensasi yang diperoleh karyawan pada Bank Perkreditan Rakyat
di Kabupaten Buleleng berpengaruh signifikan negatif terhadap kecurangan
akuntansi pada Bank Perkreditan Rakyat di Kabupaten Buleleng dengan nilai
koefisien regresi -0,161 dengan nilai signifikansi uji t sebesar 0,042.
5. Budaya organisasi yang diterapkan pada Bank Perkreditan Rakyat di Kabupaten
Buleleng berpengaruh signifikan negatif terhadap kecurangan akuntansi pada
Bank Perkreditan Rakyat di Kabupaten Buleleng dengan nilai koefisien regresi -
0,129 dengan nilai signifikansi uji t sebesar 0,033.
9. Keterbatasan Penelitian
Berdasarkan proses review yang kami lakukan, maka keterbatasan dari penelitian
ini adalah sebagai berikut.
a. Tidak terdapat kerangka penelitian dan rumusan masalah yang ditampilkan
kurang jelas.
b. Penyajian kesimpulan yang kurang lengkap.
c. Literatur review kurang jelas.
10. Kelebihan Penelitian
a. Hasil penelitian disajikan dengan baik
11. Penelitian selanjutnya
Untuk penelitian selanjutnya perlu ditambahkan metode wawancara langsung pada
masing-masing responden dalam upaya mengumpulkan data, sehingga dapat
menambah keabsahan data yang diperoleh dan lebih mendalami bahan kajian
penelitian
REVIEW ARTIKEL INTERNASIONAL
PREMA SANJAYA (1981621023)

Judul : A Hidden Risk of Auditor Industry Specialization: Evidence from the


Financial Crisis
Penulis : Cory Cassell, Emily Hunt, Gans Narayanamoorthy, dan Stephen P.
Rowe
Sumber : Review of Accounting Studies, Vol 8, tahun 2019

1. Latar Belakang
Penelitian yang luas dalam akuntansi menunjukkan bahwa auditor yang
diklasifikasikan sebagai spesialis industri memberikan audit dengan kualitas yang
lebih tinggi. Kesimpulan tentang manfaat spesialisasi industri auditor umumnya
didasarkan pada analisis yang menggunakan pengumpulan sampel dalam jumlah
besar, di mana manfaat rata-rata diukur dari waktu ke waktu. Namun, peneliti
menggunakan argumen teoritis dan bukti dari literatur keuangan untuk
mengidentifikasi situasi di mana spesialisasi industri auditor dapat merusak hasil
audit.
Penelitian ini menyelidiki kinerja auditor spesialis selama periode risiko tinggi.
Dalam lingkungan yang stabil, spesialis industri harus dapat menerapkan pengalaman
dan pengetahuan unik mereka dengan cara yang menghasilkan hasil audit yang lebih
baik. Namun, selama periode peningkatan risiko spesifik industri, spesialis dari
industri yang terkena dampak dapat berjuang untuk mengamankan dan
mengalokasikan sumber daya yang cukup untuk mengurangi risiko yang meningkat
di industri yang terkena dampak. Sebaliknya, auditor non-spesialis memiliki
portofolio klien yang kurang terkonsentrasi di industri yang terpengaruh,
menunjukkan bahwa mereka harus lebih mampu mengamankan dan mengalokasikan
sumber daya internal untuk klien di industri yang terpengaruh.
Peneliti dalam penelitian ini menyelidiki efek dari krisis keuangan pada audit
klien dari industri perbankan. Krisis keuangan dan industri perbankan memberikan
latar yang ideal untuk penyelidikan karena (1) krisis keuangan mempengaruhi hampir
semua bank dalam industri, (2) risiko nasabah bank meningkat relatif terhadap klien

24
lain selama krisis keuangan, (3) industri perbankan mewakili eksposur risiko yang
signifikan untuk banyak kantor audit, (4) variasi cross-sectional yang signifikan ada
di kantor audit dalam eksposur mereka ke industri perbankan, dan (5) audit bank
memerlukan pengetahuan khusus, menunjukkan bahwa mereka dapat mengambil
manfaat dari keahlian industri auditor. Peneliti berpendapat bahwa krisis keuangan
meningkatkan risiko klien perbankan dan pengawasan regulasi dan pasar yang
dihadapi oleh auditor eksternal bank.

2. Tujuan Penelitian
Tujuan dari penelitian ini adalah untuk mengidentifikasi situasi di mana
spesialisasi industri auditor dapat merusak hasil audit. Peneliti memperkirakan bahwa
selama periode peningkatan risiko spesifik industri, auditor spesialis dari industri
yang terpengaruh dapat berjuang untuk mengamankan dan mengalokasikan sumber
daya yang cukup untuk mengurangi risiko yang meningkat. Hal ini disebabkan karena
mereka memiliki portofolio klien yang terkonsentrasi di industri yang terpengaruh.

3. Tinjauan Literature
3.1 Spesialisasi Industri Auditor
Balsam et al. (2003) dan Krishnan (2003) menemukan bahwa klien auditor
yang merupakan spesialis industri nasional melaporkan akrual abnormal yang
lebih kecil. Reichelt dan Wang (2010) menemukan bahwa kualitas audit lebih
tinggi untuk perusahaan audit yang diklasifikasikan sebagai spesialis industri kota
dan nasional. Krishnan (2005) menemukan bahwa klien auditor spesialis
melaporkan pendapatan yang mencerminkan pengakuan berita buruk yang lebih
tepat waktu daripada klien auditor non-spesialis. Taylor (2010) menemukan
bahwa untuk akun yang sangat spesifik untuk industri perbankan, auditor spesialis
bank lebih percaya diri dalam penilaian risiko inheren mereka daripada auditor
lain.

3.2 Diversifikasi Industri Auditor


Beberapa penelitian sebelumnya berpendapat bahwa perusahaan melakukan
diversifikasi untuk meningkatkan kinerja perusahaan. Diversifikasi dikaitkan

25
dengan peningkatan kinerja perusahaan karena memungkinkan perusahaan
mengalokasikan sumber daya dengan lebih baik dan mengelola risiko.
Diversifikasi dapat bermanfaat ketika sumber daya dapat ditransfer dengan mudah
di seluruh area, seandainya risiko atau peluang signifikan muncul. Pandya dan
Rao (1998) dan Rumelt (1982) menyatakan bahwa terdapat hubungan positif
antara diversifikasi perusahaan dengan nilai perusahaan. Lang dan Stulz (1994)
menyatakan bahwa terdapat hubungan negatif antara diversifikasi perusahaan
dengan nilai perusahaan.

3.3 Risiko Audit dan Alokasi Sumber Daya


Bell et al. (2008) menyatakan bahwa auditor mengalokasikan sumber daya
yang lebih banyak dan lebih baik (yaitu jam personel) untuk klien dengan risiko
lebih tinggi. Penelitian lain telah mendokumentasikan serangkaian respons
terhadap peningkatan risiko termasuk menyesuaikan rencana audit dan biaya audit
(Pratt dan Stice, 1994; Simunic dan Stein, 1996), meningkatkan tingkat
skeptisisme profesional (Payne dan Ramsay, 2005), meningkatkan penerbitan
modifikasi pendapat (Krishnan dan Krishnan, 1996), dan mengundurkan diri dari
klien (Krishnan et al., 2013).

3.4 Mengaudit Risiko di Industri Perbankan dan Krisis Keuangan


Bajaj dan Creswell (2008) menunjukkan bahwa auditor bank menanggapi
krisis keuangan sesuai dengan standar audit saat ini dengan menyesuaikan rencana
audit dalam menanggapi risiko yang meningkat. Doogar et al. (2015)
menggunakan biaya audit sebagai proxy untuk perhatian auditor dan menemukan
bahwa perhatian auditor selama menjelang krisis keuangan bergeser sejalan
dengan berbagai sumber risiko audit.

3.5 Tekanan Beban Kerja Auditor


Bills et al. (2015) mendokumentasikan penurunan kualitas audit yang
konsisten dengan kendala sumber daya ketika perusahaan audit mengalami
pertumbuhan substansial di tingkat kantor. Coram et al. (2004) menemukan
bahwa auditor terlibat dalam tindakan pengurangan kualitas sebagai respon

26
strategis terhadap tekanan waktu. Lopez dan Peters (2012) menemukan bahwa
akrual abnormal lebih besar dan perusahaan lebih cenderung memenuhi atau
mengalahkan tolok ukur pendapatan untuk audit yang diselesaikan selama musim
sibuk audit.

4. Hipotesis
Hipotesis yang dirumuskan dalam penelitian ini adalah:
H1 : Manfaat spesialisasi industri auditor perbankan untuk hasil audit klien bank
dikurangi selama krisis keuangan, dibandingkan dengan periode sebelum
krisis.

5. Metodologi Penelitian
5.1 Variabel Penelitian
Peneliti melakukan pengujian dengan menggunakan tiga hasil audit yang
dapat diamati yang mencakup kualitas dan ketepatan waktu audit. Peneliti
menggunakan salah saji laporan keuangan untuk mewakili audit yang berkualitas
rendah, serta menggunakan penundaan opini audit dan keterlambatan pengajuan
untuk mewakili ketepatan waktu audit.

5.2 Populasi dan Sampel


Populasi dalam penelitian ini adalah semua perusahaan dari tahun 2004
hingga 2009 dengan data yang diperlukan dalam Compustat dan Audit Analytics.
Sampel dibatasi untuk klien dalam industri perbankan, diklasifikasikan sebagai
kode SIC 60 hingga 62 yang meliputi lembaga penyimpanan, lembaga
nondepositori, dan broker keamanan dan komoditas. Peneliti juga membatasi
sampel untuk bank domestik dengan auditor domestik. Peneliti mengecualikan
perusahaan audit dengan kurang dari 10 klien publik. Sampel akhir dari penelitian
terdiri dari 3.199 amatan.

27
5.3 Teknik Analisis Data
Teknik analisis data yang dilakukan dalam penelitian ini adalah regresi
Ordinary Least Square (OLS). Penelitian ini juga melakukan pengujian tambahan
dengan melakukan analisis cross-sectional.

6. Hasil Penelitian
a. Hasil analisis menunjukkan bahwa koefisien BANK_SPEC negatif dan
signifikan. Hal ini menunjukkan bahwa selama periode sebelum krisis, hasil audit
meningkat seiring dengan meningkatnya spesialisasi industri auditor perbankan.
b. Hasil analisis menunjukkan bahwa koefisien interaksi (BANK_SPEC x CRISIS)
positif dan signifikan. Hal ini menunjukkan bahwa manfaat spesialisasi industri
perbankan dikurangi selama krisis.

7. Kesimpulan
Hasil penelitian ini mendukung H1 yang menunjukkan bahwa manfaat
spesialisasi industri auditor perbankan untuk hasil audit dikurangi selama krisis
keuangan dibandingkan dengan periode sebelum krisis. Peneliti menunjukkan bahwa
selama krisis keuangan, spesialisasi industri perbankan merugikan hasil audit (dalam
hal timbulnya salah saji dan keterlambatan opini audit) atau tidak memiliki manfaat
yang terdeteksi secara statistik (dalam hal keterlambatan pengajuan). Hasil penelitian
memberikan bukti kuat yang menunjukkan bahwa meskipun spesialisasi industri
auditor dikaitkan dengan hasil audit yang meningkat selama periode yang relatif
stabil, auditor spesialis (dan klien mereka) mengalami risiko yang signifikan selama
krisis.

8. Keterbatasan Penelitian
Hasil audit yang diamati selama krisis tunggal menyebabkan hasil penelitian
tidak dapat mengeneralisasi krisis dengan besaran yang berbeda atau krisis yang
terutama memengaruhi industri lain. Selain itu, hasil penelitian tidak menunjukkan
bahwa spesialisasi industri auditor rata-rata merugikan. Hasil penelitian hanya dapat
menunjukkan bahwa spesialisasi industri auditor dapat merugikan dalam keadaan
tertentu.

28
Nama : Ni Luh Putu Ayu Lastri Pramiswari
NIM : 1981621024

Judul : The Effect of Discretionary Accruals on Financial StatementFraud: The Case


of the French Companies.
Penulis : Amara Ines
Jurnal : International Researc Journal of Finance and Economics ISSN 1450-2887
Issue 161 May, 2017

1. AREA OF INTEREST

Penelitian ini untuk memeriksa efek akrual diskresi dan mekanisme tata kelola dalam
terjadinya penipuan laporan keuangan pada perusahaan Prancis.

2. PHENOMENA
Penipuan laporan keuangan didefinisikan sebagai upaya yang disengaja oleh
perusahaan untuk menipu atau menyesatkan pengguna laporan keuangan yang
diterbitkan, terutama investor dan kreditur, dengan menyiapkan dan
menyebarluaskan stat ement keuangan yang terlewatkan secara material. Penipuan
laporan keuangan telah menarik perhatian publik, pers, investor, komunitas
keuangan, dan regulator karena banyak penipuan yang terdeteksi seperti Lucent,
Xerox, Cendant, Enron, WorldCom, Adelphia, dan Tyco. Beberapa skandal
keuangan ini menunjukkan perbedaan yang signifikan antara informasi akuntansi
yang diungkapkan oleh manajer dan realitas ekonomi perusahaan. Menurut
Association of Certified Fraud Examiners (ACFE)1, penipuan perusahaan di
seluruh dunia telah mencapai angka $ 3,5 miliar pada tahun 2011. Dalam upaya
untuk mengisi kesenjangan, kami mencoba pilihan perusahaan Prancis, kami
membenarkan pilihan menurut studi terbaru 2014 yang dilakukan oleh perusahaan
Price water house Coopers (PWC) melakukan bahwa 55% perusahaan Prancis
melaporkan telah mengalami penipuan akuntansi.
3. THEORETICAL FOUNDATION
Adapun teori yang diuji dalam penelitian ini diantaranya:
1. Akrual Diskresi
Choi et al. (2011) dan Hasnan et al. (2012), Studi menunjukkan bahwa manajer
menggunakan akrual diskresi sebagai cara untuk menanggapi informasi asimetris
dan biaya agensi dalam informasi dan strategi pribadi mereka untuk informasi
keuangan.
2. Variabel Kontrol (Berle and Means, 1932), Teori agensi menunjukkan bahwa
ada konflik kepentingan dan asimetri informasi antara manajemen dan
pemegang saham, konflik ini dapat menjelaskan penipuan perusahaan.
3. Variabel Tata Kelola Perusahaan Teori agensi menunjukkan bahwa karakteristik
dewan dapat meningkatkan kebijakan pelaporan perusahaan melalui
pengungkapan dan transparansi yang lebih sukarela (Fama dan Jensen,
1983). Ukuran papan dapat memainkan peran penting dalam mempromosikan
transparansi perusahaan dan dalam kemampuanors langsung untuk memantau
dan mengontrol manajer karena meningkatkan kumpulan keahlian dan pendapat
dan akibatnya mengurangi kemungkinan manipulasi (Jensen, 1993).
4. Karakteristik Bisnis Keuangan Karakteristik bisnis keuangan Perusahaan dapat
mencakup kinerja keuangan, rasio utang, rasio pertumbuhan dan ukuran
perusahaan. Perusahaan lebih cenderung memanipulasi angka akuntansi ketika
kinerja relatif rendah.
5. Tes Sensitivitas Piot dan Janin (2007) menggunakan nilai absolut akrual
abnormal untuk mengukur keseluruhan kecenderungan manajemen penghasilan.
Menurut Warfield dkk. (1995), akrual abnormal absolut condong untuk
menentukan tingkat ke mana manajer dengan sengaja mengejar teknik untuk
menyesuaikan angka yang dilaporkan.

4. RESEARCH QUESTION

Bagaimana pengaruh aktual diskresi terhadap penipuan laporan keuangan terhadap


kasus perusahaan di prancis?
5. METHODOLOGY
Memeriksa efek dari akrual diskresi keseluruhan pada model penipuan
akuntansi, analisis data logit dilakukan untuk sampel 250 pengamatan yang
mencakup dari 2006 hingga 2010. Model pertama kami (ditandai sebagai Model 1)
sejalan dengan penyelidikan empiriseral sevmengenai masalah yang sama termasuk
faktor-faktor yang terkait dengan motivasi dan peluang, yang mengarah pada
penipuan dalam laporan keuangan.

6. DATA AND METHOD

Sampel ini terdiri dari 250 laporan tahunan yang mencakup dari 2006 hingga 2010
untuk perusahaan Prancis terdaftar, yang 45 perusahaan fraudulent terdeteksi oleh
Otoritas Pasar Keuangan (AMF).

7. FINDINGS
Temuan ini memiliki implikasi kebijakan bagi praktisi dan regulator untuk
mengurangi kemungkinan penipuan perusahaan, mengingat peran penting yang
dimainkan oleh akrual diskresi sebagai indikator penting untuk menentukan
penipuan akuntansi. Penelitian ini adalah yang pertama di Prancis yang fokus
pada bagaimana akrual diskresi dan mekanisme tata kelola terbukti dalam
perusahaan penipuan yang dikenai sanksi oleh otoritas pasar keuangan (AMF).
Hasil pengujian hipotesis akan diuraikan pada pembahasan dibawah ini :
H1: Ada hubungan yang signifikan antara akrual diskresi dan kemungkinan
penipuan laporan keuangan.
H1.a. Akrual diskresi positif (akuntansi agresif) meningkatkan kemungkinan
penipuan laporan keuangan.
H1.b Akrual diskresi negatif (akuntansi konservatif) mengurangi kemungkinan
penipuan laporan keuangan.

8. CONCLUSIONS
Tujuan dari penelitian ini adalah untuk memeriksa efek akrual diskresi untuk
menentukan penipuan laporan keuangan, dalam hal ini, kami menguji enam
model yang memungkinkan adalah untuk mengetahui besaran akrual pada
kemungkinan menentukan penipuan akuntansi, juga kami mencatat, bahwa tanda-
tanda akrual memiliki efek penting, perusahaan terlibat lebih sering dalam akrual
diskresi positif sebelum terlibat dalam tindakan ilegal. Inilah sebabnya mengapa
studi kami sejalan dengan theresearchersof Beasley et al. (2010), Perols dan
Lougee (2011) dan Trompeter dkk. (2013)yang menunjukkan bahwa indikator
penting untuk mendeteksi penipuan adalah akrual diskresi. Kami melihat bahwa
penelitian sebelumnya secara teoritis terkait dengan penipuan akuntansi dengan
lemahnya sistem tata kelola (Hasnan et al., 2012; Kalbers, 2009; Orang-orang, 2006),
ini juga ditemukan dalam studi kami, direktur luar dan kepemilikan memiliki efek
yang signifikan pada terjadinya penipuan. Sebaliknya, sebagian besar penelitian
sebelumnya menegaskan bahwa mekanisme tata kelola tidak memiliki efek
signifikan seperti Brown dkk (2011), menjelaskan efek dari peraturan tdia dan aturan
karakteristik tata kelola adalahumum di semua perusahaan, yang membuatnya lebih
sulit untuk mengidentifikasi perusahaan yang terlibat dalam transaksi penipuan.
Selain itu, mereka berpendapat bahwa perusahaan penipuan dapat mengubah dewan
direktur mereka sebagaimanajerd. Dengan demikian, lebih sulit untuk
mengidentifikasi mereka yang kemungkinan besar terlibat dalam penipuan. Ini
menguatkan dengan studi kami, baik dewan direksi ukuran maupun CEO dualitas,
tidak memiliki efek yang signifikan dalam penipuan akuntansi.
Penelitian ini adalah yang pertama di Prancis yang fokus pada bagaimana
akrual diskresi (sinyal mutlak dan yang berbeda) dan mekanisme tata kelola terbukti
dalam perusahaan penipuan yang dikenai sanksi oleh otoritas pasar keuangan (AMF).
Menurut Delvaille et al. (2005), menyatakanbahwa o bjectives otoritas publik
adalah untuk meningkatkan fungsi sistem keuangan dan meningkatnya tingkat
investor Kepercayaan. Aliran penelitian yang besar (misalnyaHui et al. 2014,
Beneish et al. 2012, Dechow et al.2011, Jones et al. 2008, Beasley et al. 2010)
menimbulkan pertanyaan tentang informasi penipuan yang terdeteksi oleh SEC
(Komisi Sekuritas dan Bursa). Namun, sedikit perhatian telah diberikan pada
fenomena ini di tempat-tempat lain di dunia seperti Prancis. Jelas, kasus penipuan di
Amerika Serikat telah banyak dipelajari dalam jurnal akademik dan profesional.
9. FURTHER RESEARCHES
Untuk para meneliti selanjutnya di harapkan dapat memperluas sampel penelitian
sehingga hasilnya dapat diseralisasi dan juga dapat digunakan untuk menambah
atau mengganti Variabel lain yang dapat mempengaruhi penipuan laporan
keuangan pada Perusahaan.
Nama :I Wayan Megayana
NIM : 1981621004
Ringkasan Reviu Artikel Nasional

Judul : Pengaruh Audit Tenure, Audit Fee, Rotasi Auditor, Spesialiasi Auditor,
dan Umur Publikasi Pada Kualitas Audit
Penulis : I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika
Publikasi : E-Jurnal Akuntansi Universitas Udayana Vol.19.1. April (2017): 168- 194

1. Masalah Penelitian
Masa perikatan/kontrak audit atau audit tenure dapat mempengaruhi kualitas
audit. Masa perikatan audit sempat menjadi perdebatan publik. Apabila masa perikatan
yang dilakukan dalam waktu lama akan membuat kedekatan yang berlebihan antara klien
dan auditor sehingga independensi dan objektivitas auditor nantinya akan menurun.
Pemberian imbalan jasa atau fee sesuai keinginan dapat memotivasi auditor untuk
meningkatkan kinerjanya dalam mengaudit laporan perusahaan (Maharani, 2014). Jika
semakin besar imbalan yang diberikan klien kepada auditor dapat membuat auditor
memaksimalkan kemampuannya dalam melakukan audit dan memperbaiki kinerjanya
menjadi lebih baik. Jika imbalan yang diberikan kecil dapat membuat auditor menjadi
tidak termotivasi untuk meningkatkan kemampuan dan melaksanakan tugas dengan baik.
Adanya aturan dari pemerintah tentang rotasi auditor dapat berdampak baik bagi klien.
Myers (2013) menyatakan tujuan adanya rotasi auditor dalam mengaudit perusahaan
untuk mencegah klien dan auditor memiliki kedekatan yang berlebihan. Pengalaman yang
dimiliki auditor dalam melakukan audit perusahaan yang sama menjadikan auditor
tersebut memiliki spesialisasi. Auditor spesialis memiliki pengalaman audit perusahaan
yang sama sebelumnya dan lebih memahami kondisi perusahaan. Semakin singkat umur
publikasi suatu perusahaan, maka perusahaan tersebut dianggap memiliki kualitas dan
manajemen yang kurang bagus. Manajer dianggap tidak mampu menyediakan informasi
dengan benar dan menghasilkan kualitas laporan yang kurang akurat bagi pengguna
informasi.
2. Tujuan
Penelitian ini bertujuan memberikan bukti empiris tentang pengaruh audit tenure,
audit fee, rotasi auditor, spesialisasi auditor, dan umur publikasi pada kualitas audit.

3. Teori Relevan
Auditor harus memiliki motivasi yang jelas dalam melasanakan tugasnya agar
dapat mencapai tujuan audit dan tujuan perusahaan. Ada beberapa faktor yang dapat
memotivasi auditor meningkatkan kemampuannya dalam bekerja seperti adanya imbalan
yang besar dan pemberian hadiah atau penghargaan (Tan, 2000). Auditor yang
termotivasi untuk mendapatkan imbalan yang besar dari klien akan berusaha menujukkan
kemampuan terbaik yang dimilikinya serta menghasilkan kualitas yang baik.

4. Hipotesis
H1 : Audit tenure berpengaruh negatif pada kualitas audit.
H2 : Audit fee berpengaruh positif pada kualitas audit
H3 : Rotasi auditor berpengaruh positif pada kualitas audit
H4 : Spesialisasi auditor berpengaruh positif pada kualitas audit
H5 : Umur publikasi berpengaruh positif pada kualitas audit.

5. Variabel
a. Variabel Independen : Audit tenure, Audit fee, Rotasi auditor, Spesialisasi auditor,
Umur publikasi
b. Variabel Dependen : Kualitas audit

6. Data
Adapun yang Populasi dalam penelitian ini yaitu seluruh perusahaan manufaktur yang
tercatat di bursa efek Indonesia selama periode 2013-2015. Sampel yang digunakan hanya
beberapa dari keseluruhan jumlah perusahaan yang ada. Metode yang digunakan untuk
menentukan sampel adalah purposive sampling yang berisi kriteria tertentu. Metode yang
digunakan untuk mengumpulkan data adalah metode observasi non partisipan yang mana
peneliti tidak harus terjun langsung ke lapangan untuk mendapatkan data.
7. Uji Statistik
Adapun dalam teknik analisis menggunakan regresi logistik karena menggunakan
variabel dummy pada variabel terikatnya. Ada beberapa langkah yang harus dilakukan
sebelum melakukan analisis data yaitu menilai kelayakan model, menilai keseluruhan
model, menguji koefisien determinasi, menguji multikolinearitas, menguji matriks
klasifikasi, dan mendapatkan model regresi yang terbentuk (Ghozali, 2012).

8. Hasil

a. Hasil uji H1 menunjukkan nilai signifikansi tenur sebesar 0,000 dan kurang dari
0,05. Hal ini membuktikan hipotesis dapat diterima yaitu audit tenure mempunyai
pengaruh negatif pada kualitas audit.
b. Hasil uji H2 menunjukkan nilai signifikansi fee 0,000 dan kurang dari 0,05. Hal ini
membuktikan hipotesis dapat diterima yaitu audit fee mempunyai pengaruh positif
pada kualitas audit.
c. Hasil uji H3 menunjukkan nilai signifikansi sebesar 0,140 yang lebih besar dari
0,05. Hal ini membuktikan hipotesis ditolak yaitu rotasi auditor tidak mempunyai
pengaruh pada kualitas audit.
d. Hasil uji H4 menunjukkan nilai signifikansi 0,997 dan lebih dari 0,05. Hal ini
membuktikan hipotesis ditolak yaitu spesialisasi auditor tidak memberikan
pengaruh pada kualitas audit.
e. Hasil uji H5 menunjukkan nilai signifikansi sebesar 0,798 yang lebih besar dari
0,05. Hal ini membuktikan hipotesis ditolak yaitu umur publikasi tidak berpengaruh
pada kualitas audit.

9. Kesimpulan
Dari hasil penelitian, maka dapat disimpulkan audit tenure dapat memberi pengaruh
negatif, audit fee dapat memberi pengaruh positif, umur publikasi, rotasi auditor, dan
spesialisasi auditor tidak memberikan pengaruh pada kualitas audit perusahaan
manufaktur di Bursa Efek Indonesia periode 2013-2015.
10. Saran
Berdasarkan pada simpulan, peneliti berikutnya dapat menambahkan jumlah periode
pengamatan dalam riset yaitu selama lebih dari 3 tahun, menambahkan beberapa variabel
bebas untuk menyempurnakan penelitian ini, dan memperluas sektor industri yang
dijadikan populasi agar mendapat hasil yang pasti.
REFERENSI

Amara Ines. The Effect of Discretionary Accruals on Financial Statement Fraud: The
Case of the French Companies. International Researc Journal of Finance and
Economics ISSN 1450-2887 Issue 161 May, 2017.
Cory Cassell, Emily Hunt, Gans Narayanamoorthy, dan Stephen P. Rowe. A Hidden Risk
of Auditor Industry Specialization: Evidence from the Financial Crisis. Review of
Accounting Studies, Vol 8, tahun 2019.
Ketut Budiartini, Gusti Ayu Rencana Sari Dewi, Nyoman Trisna Herawati. Faktor-Faktor
Yang Mempengaruhi Terjadinya Kecurangan Akuntansi Dalam Perspektif Fraud
Diamond (Studi Empiris pada Bank Perkreditan Rakyat (BPR) Kabupaten
Buleleng). Jurnal Ilmiah Mahasiswa Akuntansi Universitas Pendidikan Ganesha,
Vol: 10 No: 2 Tahun 2019 e-ISSN: 2614 – 1930
Michael C. Knapp. 2007. Auditing Cases. Sixth Edition. International Student Edition.
South-Western: Thomson
Pramaswaradana I Gusti Ngurah Indra, Astika Ida Bagus Putra. Pengaruh Audit Tenure,
Audit Fee, Rotasi Auditor, Spesialsiasi Auditor, Dan Umur Publikasi Pada
Kualitas Audit. E-Jurnal Akuntansi Universitas Udayana Vol.19.1. April (2017):
168-194.
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun
2019 e-ISSN: 2614 – 1930

FAKTOR-FAKTOR YANG MEMPENGARUHI TERJADINYA


KECURANGAN AKUNTANSI DALAM PERSPEKTIF
FRAUD DIAMOND
(Studi Empiris pada Bank Perkreditan Rakyat (BPR)
Kabupaten Buleleng)
Ketut Budiartini1
Gusti Ayu Rencana Sari Dewi1, Nyoman Trisna Herawati2

Jurusan Akuntansi Program S1 Fakultas Ekonomi


Universitas Pendidikan Ganesha
Singaraja, Indonesia
e-mail: {budiartini538@gmail.comayurencana@undiksha.ac.id
aris_herawati@yahoo.co.id}
ABSTRAK

Penelitian ini bertujuan untuk menguji pengaruh dari elemen fraud diamond yaitu
pengaruh 1) posisi jabatan, 2) keefektifan pengendalian internal, 3) penegakan
peraturan, 4) kesesuaian kompensasi, 5) budaya organisasi terhadap kecurangan
akuntansi. Jenis penelitian ini adalah penelitian kuantitatif. Populasi dalam penelitian ini
adalah seluruh karyawan Bank Perkreditan Rakyat (BPR) di Kabupaten Buleleng.
Pengambilan sampel menggunakan metode non probability sampling dengan teknik
purposive sampling, dengan responden teller, collector, account officer, marketing, back
office dan admin kredit. Teknik pengumpulan data yang digunakan terdiri dari 6
instrumen dengan menggunakan skala likert. Uji kualitas data yang digunakan yaitu uji
validitas dan reliabilitas. Uji asumsi klasik yang digunakan yaitu uji normalitas,
multikolinearitas, serta heteroskedastisitas. Metode analisis data yang digunakan adalah
analisis regresi linier berganda dengan bantuan SPSS 24.0 for Windows.Kesimpulan
penelitian menunjukkan: 1) Posisi jabatan berpengaruh signifikan positif terhadap
kecurangan akuntansi, 2) Keefektifan pengendalian internal, penegakan peraturan,
kesesuaian kompensasi dan budaya organisasi berpengaruh signifikan negatif terhadap
kecurangan akuntansi.
Kata kunci:Budaya, Jabatan, Kecurangan, Kompensasi, Pengendalian Internal,
Peraturan

ABSTRACT

This research aimed at examining the effect of diamond fraud elements, namely
the effect of 1) position at the office, 2) internal control effectiveness, 3) regulations
enforcement, 4) compensation suitability, 5) organizational culture on accounting fraud.
This type of research was a quantitative research. The population in this study was all
employees of the Rural Bank (BPR) in Buleleng Regency. The sampling employed non
probability sampling method with purposive sampling technique, with the respondents of
tellers, collectors, account officers, marketing, back officers, and credit administrators.
The data collection technique used consisted of 6 instruments using a Likert scale. The
tests of data quality used were validity and reliability test. The classic assumption test
used was normality test, multicollinearity, and heteroscedasticity. The data analysis
method used was multiple linear regression analysis through SPSS 24.0 for Windows.
The conclusion of the research showed: 1) position at the office had a positive significant
effect on accounting fraud, 2) internal control effectiveness, regulations enforcement,
compensation suitability, and organizational culture had a significant negative effect on
accounting fraud.
Keywords: Culture, Position, Fraud, Compensation, Internal Control, Regulation

113
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun 2019 e-ISSN:
2614 – 1930

PENDAHULUAN nya, penelitian ini berusaha untuk menggali


persepsi para karyawan untuk mengetahui
Kecurangan (fraud) adalah salah satu faktor-faktor yang menyebabkan terjadinya
penipuan yang sengaja dilakukan sehingga kecurangan (fraud) di sektor perbankan
dapat menimbulkan kerugian tanpa disadari khususnya di Bank Perkreditan Rakyat
oleh pihak yang dirugikan tersebut dan (BPR) Kabupaten Buleleng dengan meng-
memberikan keuntungan bagi pelaku gunakan konsep fraud diamond.
kecurangan. Singelton dan Aaron J. (2010), Elemen dalam konsep fraud diamond
mendefinisikan kecurangan (fraud) sebagai ini kemudian diproksikan dengan variabel-
perbuatan yang mencakup akal muslihat, variabel agar lebih mudah diukur, yaitu
kelicikan, dan tidak jujur dan cara-cara tekanan (pressure) yang diproksikan
yang tidak layak/wajar untuk menipu orang dengan variabel kesesuaian kompensasi;
lain untuk keuntungan diri sendiri, sehingga variabel kesempatan (opportunity) yang
menimbulkan kerugian bagi pihak lain. diproksikan dengan variabel keefektifan
Dampak yang dapat diakibatkan oleh pengendalian internal, dan penegakan
adanya kecurangan (fraud) antara lain, peraturan; variabel rasionalisasi (ratio-
hancurnya reputasi organisasi, kerugian nalization) yang diproksikan dengan vari-
organisasi, rusaknya moralitas karyawan abel budaya organisasi; serta variabel ke-
serta dampak-dampak negatif lainnya. mampuan (capability) yang diproksikan
Kecurangan akuntansi telah terjadi di dengan posisi jabatan dalam perusahaan.
berbagai sektor perbankan. Bahkan di Bali Peneliti memilih Bank Perkreditan Rakyat
telah terjadi beberapa kasus terkait (BPR) Kabupaten Buleleng sebagai lokasi
kecurangan akuntansi di Bank Perkreditan penelitian dengan alasan adanya beberapa
Rakyat (BPR). Ketidaksehatan BPR ini kasus fraud yang terjadi di Bank Per-
disebabkan oleh tingginya tingkat ke- kreditan Rakyat (BPR) Kabupaten Buleleng
curangan yang dilakukan oleh internal BPR tersebut selama beberapa tahun terakhir
itu sendiri. Salah satu kasus yang sangat dimana pelaku fraud lebih banyak dilakukan
ramai diperbincangkan saat ini adalah oleh orang yang memiliki jabatan pekerjaan
kasus penggelapan dana nasabah yang yang tinggi dalam perusahaan dan belum
dilakukan oleh salah satu karyawan pada ada yang melakukan penelitian terkait
PD. BPR Bank Buleleng 45 Kas Seririt yang penyebab kecurangan yang terjadi pada
menelan kerugian senilai lebih dari Rp 635 Bank Perkreditan Rakyat (BPR) tersebut.
juta (Singaraja, Balipost.com). Kemudian Tujuan dari adanya penelitian ini
kasus penggelapan dana yang terjadi pada adalah untuk mengetahui pengaruh dari
tahun 2015 adalah kasus yang dilakukan posisi jabatan, keefektifan pengendalian
oleh Kolektor PT. BPR Suryajaya internal, penegakan peraturan, kesesuaian
Kubutambahan, Kecamatan Kubutambah- kompensasi dan budaya organisasi ter-
an, Kabupaten Buleleng, Bali, Jhonieko hadap kecurangan akuntansi pada Bank
Maralia Siden alias Joni. Ia diamankan Perkreditan Rakyat (BPR) di Kabupaten
karena terbukti menggelapkan (menilep) Buleleng. Jika sudah diketahui pengaruh-
uang nasabah BPR senilai Rp 200 juta nya maka hal itu dapat dijadikan suatu
(Bali, Tribunnews.com). acuan bagi Bank Perkreditan Rakyat (BPR)
Adanya beberapa kasus yang terjadi di Kabupaten Buleleng untuk menghindari
di Bank Perkreditan Rakyat (BPR) serta terjadinya kecurangan akuntansi di masa
masih terbatasnya penelitian yang meng- mendatang.
gunakan elemen fraud diamond untuk me- Terdapat beberapa penelitian yang
ngetahui faktor-faktor yang mempengaruhi dilakukan untuk mengetahui maupun meng-
terjadinya kecurangan di sektor perbankan analisis penyebab kecurangan dengan
khususnya di Bank Perkreditan Rakyat menggunakan konsep fraud diamond.
(BPR), membuat peneliti tertarik untuk Salah satunya yaitu penelitian yang dilaku-
melakukan penelitian dengan mengguna- kan oleh I.G.A. Ngurah Rahma Y. (2017),
kan elemen-elemen dalam konsep fraud posisi jabatan merupakan hak yang formal
diamond. Dengan mereplikasi dan dan sah untuk mengeluarkan wewenang
memodifikasi penelitian-penelitian sebelum- dalam mengambil keputusan, mengeluar-

114
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun 2019 e-ISSN:
2614 – 1930

kan perintah, dan mengalokasikan sumber H3 : Penegakan Peraturan ber-


daya agar tercapai. pengaruh signifikan negatif terhadap
Terkadang posisi jabatan yang dimiliki kecurangan akuntansi
digunakan secara tidak bijak untuk mem-
peroleh keuntungan individu. Jadi semakin Kecurangan akuntansi juga dapat
tinggi posisi jabatan yang diperoleh maka terjadi karena kompensasi yang didapatkan
semakin besar pula tindakan kecurangan karyawan tidak sesuai dengan yang se-
(fraud) yang dapat terjadi. Berdasarkan harusnya. Menurut penelitian Prekanida
pemaparan di atas, dapat ditarik hipotesis (2015), kompensasi merupakan hal yang
sebagai berikut: berpengaruh terhadap perilaku karyawan,
H1 : Posisi jabatan berpengaruh seseorang cenderung akan berperilaku
signifikan positif terhadap kecurangan tidak etis dan berlaku curang untuk
akuntansi memaksimalkan keuntungan untuk dirinya
sendiri. Dengan kompensasi yang sesuai
Selain posisi jabatan yang tinggi, diharapkan dapat meminimalisasi tindakan
kecurangan akuntansi dapat terjadi apabila kecurangan akuntansi. Berdasarkan pema-
pengendalian internal dalam perusahaan paran di atas maka dapat ditarik hipotesis
tidak berjalan secara efektif. Karena sebagai berikut:
pengendalian internal tidak berjalan secara H4 : Kesesuaian Kompensasi ber-
efektif tersebut, maka peluang seseorang pengaruh signifikan negatif terhadap
untuk melakukan tindakan kecurangan kecurangan akuntansi
akuntansi sangat terbuka.
Menurut hasil penelitian yang Selain itu budaya organisasi juga
dilakukan Vani Adelin (2013); Muammar dapat mempengaruhi terjadinya kecurang-
Rizky dan Fauziah Aida Fitri (2017), an akuntansi. Menurut Moeljono dalam
keefektifan pengendalian internal dapat Zulkarnain (2013), budaya organisasi
mencegah dan mengurangi tindakan adalah sistem nilai-nilai yang diyakini
Kecurangan akuntansi dalam manajemen semua anggota organisasi dan yang
yang berarti bahwa semakin efektif dipelajari, diterapkan serta dikembangkan
pengendalian internal dalam manajemen secara berkesinambungan, yang dapat
maka semakin rendah tindakan kecurangan dijadikan acuan berperilaku dalam organi-
akuntansinya. Maka dari itu dapat ditarik sasi untuk mencapai tujuan yang telah
hipotesis sebagai berikut: ditetapkan. Kinerja karyawan dapat
H2 : Keefektifan Pengendalian Internal ditingkatkan dengan menciptakan budaya
berpengaruh signifikan negatif terhadap yang baik dan dapat digunakan untuk
kecurangan akuntansi mencapai tujuan perusahaan. Berdasarkan
pemaparan di atas maka dapat ditarik
Faktor lain yang mempengaruhi ter- hipotesis sebagai berikut:
jadinya kecurangan yaitu penegakan pe- H5 : Budaya Organisasi berpengaruh
raturan yang masih lemah. Menurut signifikan negatif terhadap kecurangan
Asshiddiqie dalam Muammar Rizky dan akuntansi
Fauziah Aida Fitri (2017), penegakan
peraturan yang baik juga akan berpengaruh METODE PENELITIAN
terhadap pelaksanaan kegiatan-kegiatan Penelitian ini dilakukan dengan
dalam perusahaan yang berhubungan metode penelitian kuantitatif. Variabel yang
dengan akuntansi dengan baik dan benar digunakan dalam penelitian ini yaitu posisi
sehingga nantinya menghasilkan kinerja jabatan, keefektifan pengendalian internal,
yang efektif dan mampu memberikan hasil penegakan peraturan, kesesuaian kompen-
yang baik untuk pihak-pihak yang ber- sasi, budaya organisasi dan kecurangan
kepentingan. Berdasarkan pemaparan di akuntansi. Penelitian ini dilakukan pada
atas maka dapat ditarik hipotesis sebagai Bank Perkreditan Rakyat (BPR) di Kabu-
berikut: paten Buleleng. Jumlah BPR yangditeliti
adalah 17 BPR dengan jumlah responden
95 orang. Teknik pengambilan sampel

115
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun 2019 e-ISSN:
2614 – 1930

dalam penelitian ini menggunakan metode pertanyaan pada kuesioner digunakan


non probability sampling denganteknik analisis item yang mengkorelasikan skor
purposive sampling dengan beberapa tiap butir dengan skor total yang
kriteria yaitu (1) Jabatan sebagai Teller, merupakan jumlah setiap skor butir.
Collector, Account Officer, Marketing, Back Berdasarkan hasil uji validitas, item
Office,dan Admin Kredit. (2) Masa kerja pernyataan variabel independen dan
minimal 3 tahun. (3) Pendidikan terakhir dependen memiliki nilai r hitung lebih besar
minimal SMA/SMK/Sederajat. dari r tabel (rhitung> rtabel). Jadi dapat
Pengumpulan data dilakukan dengan dikatakan bahwa pernyataan dalam
menyebar kuesioner sebanyak 95 eksem- kuesioner dengan variabel posisi jabatan
plar dan yang kembali 100% namun yang (X1), keefektifan pengendalian internal
tidak diisi sebanyak 11 eksemplar sehingga (X2), penegakan peraturan (X3),
kuesioner yang dapat digunakan sebanyak kesesuaian kompensasi (X4), budaya
84 eksemplar (88%). organisasi (X5) dan kecurangan akuntansi
Setelah data terkumpul maka akan (Y) adalah valid.
dilakukan uji kualitas data dengan meng- Setelah dilakukan pengujian validitas
gunakan uji validitas data dan uji reliabilitas. maka akan dilakukan uji reliabilitas. Relia-
Setelah uji kualitas data maka data akan bilitas instrumen penelitian dinilai melalui
diuji menggunakan uji asumsi klasik yang besaran koefisien Alpha Cronbach,yang
terdiri dari uji normalitas, uji multikolinea- menun-jukkan konsistensi internal item-item
ritas dan uji heteroskedastisitas. Setelah itu yang mendasari sebuah variabel.
analisis data dilakukan dengan meng- Berdasarkan hasil uji reliabilitas
gunakan teknik analisis regresi linier semua variabel memiliki Alpha Cronbach
berganda dan menggunakan uji t dan uji lebih besar dari 0,70. Jadi, dapat dikatakan
koefisien determinasi untuk mengetahui bahwa kuesioner posisi jabatan, keefektifan
pengaruh variabel bebas terhadap variabel pengendalian internal, penegakan pera-
terikat secara parsial. Pengujian ini turan, kesesuaian kompensasi, budaya
dilakukan dengan bantuan SPSS 24.0 for organisasi, dan kecurangan akuntansi
Windows. adalah reliabel.
Setelah dilakukan uji kualitas data,
HASIL DAN PEMBAHASAN maka selanjutnya dilakukan uji asumsi
Pengujiandata yang dilakukan klasik yang terdiri dari uji normailitas, uji
pertama adalah uji kualitas data dengan uji multikolinearitas, dan uji heteroskedas-
validitas dan uji reliabilitas data. Uji validitas tisitas. Uji normalitas data disajikan pada
dilakukan dengan menguji setiap butir tabel 1.

Tabel 1. Uji Normalitas Data

One-Sample Kolgomorov-Smirnov Test

Unstandardized Residual
N 84
Normal Parametersa,b Mean 0,0000000
Std. Deviation 0,59470445
Most Extreme Differences Absolute 0,057
Positive 0,039
Negative -0,057
Kolmogorov-Smirnov Z 0,057
Asymp. Sig. (2-tailed) 0,200
a. Test distribution is Normal.
b. Calculated from data.
Sumber: data diolah (2019)

116
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun 2019 e-ISSN:
2614 – 1930

Berdasarkan hasil uji normalitas data tidaknya hubungan antara variabel bebas
di atas, ditunjukkan bahwa nilai Asymp. Sig. yang satu dengan variabel yang lainnya.
(2-tailed) sebesar 0,200. Nilai Asymp. Sig. Model regresi sebaiknyanya tidak memiliki
(2-tailed) tersebut lebih besar dari 0,05 korelasi yang tinggi diantara variabel bebas
untuk statistik Kolmogorov-Smirnov Z. Hal lainnya. Uji multikolinieritas dapat diuji
ini menunjukkan bahwa sebaran data dengan menggunakan Variance Inflation
berdistribusi normal. Factor (VIF).
Langkah selanjutnya yaitu melakukan Hasil uji multikolinieritas dapat dilihat
uji multikolinearitas untuk mengetahui ada pada tabel 2 berikut.

Tabel 2. Uji Multikolinieritas

Collinearity
Model Statistics
Keterangan
Toleranc
VIF
e
Posisi jabatan 0,508 1,970 Tidak ada multikolinieritas
Keefektifan pengendalian 0,270 3,708 Tidak ada multikolinieritas
internal
Penegakan peraturan 0,391 2,556 Tidak ada multikolinieritas
Kesesuaian kompensasi 0,179 5,599 Tidak ada multikolinieritas
Budaya organisasi 0,233 4,287 Tidak ada multikolinieritas
Sumber: data diolah (2019)

Berdasarkanhasil uji multikolinearitas terjadi multikolinearitas pada model regresi


pada tabel 4, diketahui bahwa nilai VIF dari linier.Langkah selanjutnya yaitu melakukan
masing-masing variabel bebas lebih kecil uji heteroskedastisitas yang bertujuan untuk
dari 10 dan nilai tolerance lebih dari 0,10. menguji apakah dalam model regresi terjadi
Nilai korelasi di antara variabel bebas dapat ketidaksamaan varian dari residual satu
dikatakan mempunyai korelasi yang lemah. pengamatan ke pengamatan yang lain.
Jadi dapat disimpulkan bahwa antara Hasil uji heteroskedastisitas ini dapat
variabel bebas tidak ada korelasi atau tidak dilihat pada tabel 3 berikut.

Tabel 3. Uji Heteroskedastisitas

Unstandardized Standardized
Model Coefficients Coefficients t Sig.
B Std. Error Beta
1 (Constant) 1,160 1,256 0,924 0,358
X1 -0,026 0,035 -0,115 -0,744 0,459
X2 0,045 0,034 0,276 1,309 0,195
X3 -0,021 0,039 -0,096 -0,549 0,584
X4 0,027 0,047 0,152 0,585 0,560
X5 -0,057 0,036 -0,359 -1,582 0,118
a. Dependent Variable: ABS
Sumber: data diolah (2019)

Keterangan: X1 = posisi jabatan, X2 = Berdasarkanhasil uji heteros-


keefektifan pengendalian internal, dan X3 = kedastisitas di atas, diketahui bahwa nilai
penegakan peraturan, X4 = kesesuaian signifikansi antara variabel bebas dengan
kompensasi, dan X5 = budaya organisasi. absolute residual (ABS) lebih besar dari
0,05. Jadi, dapat disimpulkan bahwa model

117
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun 2019 e-ISSN:
2614 – 1930

regresi yang digunakan tidak terdapat Berdasarkan hasil analisis regresi


adanya gejala heteroskedastisitas. linier berganda dapat diketahui persamaan
Setelah melakukan uji asumsi klasik, garis regresi dengan menggunakan analisis
langkah selanjutnya adalah melakukan konstanta dan koefisien beta. Hasil
pengujian hipotesis dengan teknik analisis perhitungan konstanta dan koefisien beta
linier berganda dengan menggunakan uji t serta hasil uji t dapat dilihat pada tabel 4.
dan uji koefisien determinasi.

Tabel 4. Analisis Koefisien Beta dan Uji t

Unstandardized Standardized
Model Coefficients Coefficients t Sig.
B Std. Error Beta
1 (Constant) 20,316 2,092 9,711 0,000
X1 0,124 0,058 0,149 2,122 0,037
X2 -0,130 0,057 -0,218 -2,268 0,026
X3 -0,140 0,065 -0,171 -2,148 0,035
X4 -0,161 0,078 -0,245 -2,072 0,042
X5 -0,129 0,060 -0,224 -2,172 0,033
a. Dependent Variable: Y
Sumber: data diolah (2019)

Keterangan: X1 = posisi jabatan, X2 = akuntansi (Y) sebesar 0,124 dengan


keefektifan pengendalian internal, X3 = asumsi variabel independen yang
penegakan peraturan, X4 = kesesuaian lainnya tetap.
kompensasi, X5 = budaya organisasi, dan Y 3. Keefektifan pengendalian internal (X2)
= kecurangan akuntansi memiliki koefisien regresi -0,130. Nilai
Berdasarkan perhitungan regresi koefisien regresi yang negatif
linier berganda pada tabel 6, maka didapat menunjukkan bahwa keefektifan
hasil persamaan regresi sebagai berikut. pengendalian internal (X2) ber-
Y= α + β1X1 + β2X2 + β3X3 + β4X4 + β5X5 + ɛ pengaruh negatif terhadap kecurang-
Y = 20,316+ 0,124X1 -0,130X2 - 0,140X3 - an akuntansi (Y). Hal ini meng-
0,161X4 - 0,129X5 + ɛ gambarkan bahwa setiap kenaikan 1
Berdasarkan model regresi yang satuan keefektifan pengendalian
terbentuk, dapat diinterpretasikan hasil internal (X2) dapat menurunkan
sebagai berikut. kecurangan akuntansi (Y) sebesar
1. Konstanta 20,316 menunjukan jika 0,130 dengan asumsi variabel
variabel posisi jabatan (X1), independen yang lainnya tetap.
keefektifan pengendalian internal (X2), 4. Penegakan peraturan (X3) memiliki
penegakan peraturan (X3), nilai koefisien regresi-0,140. Nilai
kesesuaian kompensasi (X4), dan koefisien regresi yang negatif
budaya organisasi (X5) bernilai menunjukkan bahwa penegakan
konstan, maka variabel kecurangan peraturan (X3) berpengaruh negatif
akuntansi (Y) memiliki nilai 20,316 ter-hadap kecurangan akuntansi (Y).
satuan. Hal ini menggambarkan bahwa setiap
2. Posisi jabatan (X1) memiliki koefisien kenaikan 1 satuan penegakan
regresi 0,124. Nilai koefisien regresi peraturan (X3) dapat menurunkan
yang positif menunjukkan bahwa kecurangan akuntansi (Y) sebesar
posisi jabatan (X1) berpengaruh positif 0,140 dengan asumsi variabel
terhadap kecurangan akuntansi (Y). independen yang lainnya tetap.
Hal ini menggambarkan bahwa setiap 5. Kesesuaian kompensasi (X4) memiliki
kenaikan 1 satuan posisi jabatan (X1) koefisien regresi-0,161. Nilai koefisien
dapat meningkatkan kecurangan regresi yang negatif menunjukkan

118
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun 2019 e-ISSN:
2614 – 1930

bahwa kesesuaian kompensasi (X4) akuntansi (Y) sebesar 0,129 dengan


berpengaruh negatif terhadap asumsi variabel inde-penden yang
kecurangan akuntansi (Y). Hal ini lainnya tetap.
menggambarkan bahwa setiap
kenaikan 1 satuan kesesuaian kom- Tahap selanjutnya adalah melakukan
pensasi (X4) dapat menurunkan uji koefisien determinasi. Hal ini dilakukan
kecurangan akuntansi (Y) sebesar untuk mengetahuiseberapa besar pengaruh
0,161 dengan asumsi variabel variabel bebas (X) terhadap variabel terikat
independen yang lainnya tetap. (Y), yang ditunjukkan dengan nilai Adjusted
6. Budaya organisasi (X5) memiliki nilai R Square.
koefisien regresi sebesar -0,129. Nilai Pada penelitian ini menggunakan nilai
koefisien regresi yang negatif dari Adjusted R Squarekarena Adjusted R
menunjukkan bahwa budaya Squareadalah koefisien determinasi yang
organisasi (X5) berpengaruh negatif disesuaikan untuk regresi yang
terhadap kecurangan akuntansi (Y). menggunakan 3 atau lebih variabel
Hal ini menggambarkan bahwa setiap independen (Ghozali, 2011). Hasil analisis
kenaikan 1 satuan budaya organisasi koefisien determinasi dapat disajikan pada
(X5) dapat menurunkan kecurangan tabel 5 berikut.

Tabel 5. Uji Koefisien Determinasi

Model Adjusted R Std. Error of the


R R Square
Square Estimate
1 0,898 0,806 0,793 0,613
a. Predictors: (Constant), X5, X1, X3, X2, X4
Sumber: data diolah (2019)

Berdasarkanhasil uji koefisien deter- jabatan semakin tinggi, maka kecurangan


minasi pada tabel 7 di atas, diketahui akuntansi juga semakin tinggi.
bahwa koefisien determinasi sebesar 0,793. Hasil pengujian hipotesis pertama
Hal ini menunjukkan bahwa 79,3% variabel sejalan dengan teori yang dikemukakan
kecurangan akuntansi dipengaruhi oleh oleh Wolfe dan Hermanson (2004) yang
variabel posisi jabatan, keefektifan menyatakan bahwa posisi seseorang atau
pengendalian internal, penegakan pera- fungsi dalam organisasi dapat memberikan
turan, kesesuaian kompensasi, dan budaya kemampuan untuk membuat atau meman-
organisasi. Jadi disimpulkan bahwa se- faatkan kesempatan untuk penipuan. Selain
besar 20,7% dipengaruhi oleh faktor lain itu menurut Beasley dalam Ardi Iphani
yang tidak diteliti dalam penelitian ini. (2017) yang menyatakan bahwa penyalah-
gunaan aset berhubungan dengan pe-
PEMBAHASAN nyalahgunaan kekuasaan serta pengaruh
jabatan atau kedudukan istimewa dalam
Pengaruh Posisi Jabatan terhadap masyarakat untuk maksud pribadi.
Kecurangan Akuntansi Maka dapat disimpulkan bahwa posisi
Hasilpengujianhipotesispertamameng jabatan merupakan kedudukan yang dimiliki
enaipengaruh posisi jabatan ter-hadap seseorang yang memiliki wewenang yang
kecurangan akuntansi menunjukkan nilai cukup penting sehingga mudah baginya
koefisien regresi 0,124 dengan nilai untuk mengambil keputusan dengan atau
signifikansi uji t sebesar 0,037. Oleh karena tanpa persetujuan bawahannya.
itu, hipotesis pertama dalam penelitian ini Hasil penelitian I.G.A. Ngurah Rahma
diterima. Hal ini menunjukkan bahwa posisi Y. (2017) dimana posisi jabatan merupakan
jabatan berpengaruh positifsignifikan ter- hak yang formal dan sah untuk mengeluar-
hadap kecurangan akuntansi. Jika posisi kan wewenang dalam mengambil kepu-

119
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun 2019 e-ISSN:
2614 – 1930

tusan, mengeluarkan perintah, dan meng- kecurangan akuntansi dalam manajemen


alokasikan sumber daya agar tercapai. yang berarti bahwa semakin efektif
Terkadang posisi jabatan yang di-miliki pengendalian internal dalam manajemen
digunakan secara tidak bijak untuk maka semakin rendah tindakan kecurang-
memperoleh keuntungan individu. Jadi an akuntansinya.
semakin tinggi posisi jabatan yang di- Selain didukung oleh penelitian Vani
peroleh maka semakin besar pula tindakan Adelin (2013), hasil pengujian ini juga
kecurangan (fraud) yang dapat terjadi. didukung dengan hasil penelitian yang
dilakukan oleh Muammar Rizky dan
Pengaruh Keefektifan Pengendalian Fauziah Aida Fitri (2017) yang juga
Internal terhadap Kecurangan Akuntansi menyatakan bahwa pengendalian internal
Hasilpengujianhipotesiskedua me- yang efektif berpengaruh terhadap ke-
ngenaipengaruh keefektifan pengendalian curangan. Jadi semakin efektif pengen-
internal terhadap kecurangan akuntansi dalian internal suatu perusahaan maka
menunjukkan nilai koefisien regresi -0,130 tingkat terjadinya kecurangan akuntansi
dengan nilai signifikansi uji t sebesar 0,026. juga akan berkurang.
Oleh karena itu, hipotesis kedua dalam
penelitian ini diterima. Hal ini menunjukkan Pengaruh Penegakan Peraturan ter-
bahwa keefektifan pengendalian internal hadap Kecurangan akuntansi
berpengaruh negatifsignifikan terhadap Hasilpengujianhipotesisketiga me-
kecurangan akuntansi. Jika keefektifan pe- ngenaipengaruh penegakan peraturan
ngendalian internal semakin baik, maka terhadap kecurangan akuntansi menunjuk-
kecurangan akuntansi semakin rendah. kan nilai koefisien regresi -0,140 dengan
Teori yang sejalan dengan hasil nilai signifikansi uji t sebesar 0,035. Oleh
pengujian hipotesis kedua yaitu pengung- karena itu, hipotesis ketiga dalam penelitian
kapan menurut PP No. 8 Tahun 2006 ini diterima. Hal ini menunjukkan bahwa
dalam Adelin (2013) dimana pengendalian penegakan peraturan berpengaruh
internal adalah suatu proses yang negatifsignifikan terhadap kecurangan
dipengaruhi oleh manajemen yang akuntansi. Jika penegakan peraturan
diciptakan untuk memberikan keyakinan semakin baik, maka kecurangan akuntansi
yang memadai dalam pencapaian efek- semakin rendah.
tivitas, efisiensi, ketaatan terhadap pera- Hasil pengujian hipotesis ketiga se-
turan perundang-undangan yang berlaku jalan dengan teori yang dikemukakan oleh
dan keandalan penyajian laporan Asshiddiqie dalam Muammar Rizky dan
keuangan. Keefektifan pengendalian inter- Fauziah Aida Fitri (2017) yang mendefinisi-
nal dapat diartikan bahwa tujuan peru- kan penegakan peraturan sebagai proses
sahaan telah berjalan sesuai dengan dilaksanakannya upaya untuk berfungsinya
perencanaan pengendalian dan pengawas- atau tegaknya norma-norma hukum yang
an, sehingga dengan adanya pengendalian berlaku secara nyata sebagai dasar atau
dan pengawasan kecurangan tidak akan pedoman perilaku dalam kehidupan sehari-
terjadi. hari. Meningkatkan penegakan peraturan
Kecurangan akuntansi dapat terjadi pada sebuah perusahaan dapat mencegah
apabila pengendalian internal dalam peru- terjadinya kecurangan akuntansi, yang
sahaan tidak berjalan secara efektif. Untuk nantinya akan membantu perusaha-an
menutup peluang terjadinya kecurangan untuk menyediakan informasi terkait peru-
akuntansi maka suatu perusahaan harus sahaan secara objektif kepada pihak yang
memberlakukan pengendalian internal berkepentingan.Penegakan peraturan yang
secara efektif. baik juga akan berpengaruh terhadap
Hasil pengujian ini juga didukung pelaksanaan kegiatan-kegiatan dalam
dengan hasil penelitian yang dilakukan oleh perusahaan yang berhubungan dengan
Vani Adelin (2013) yang menyatakan akuntansi dengan baik dan benar sehingga
bahwa keefektifan pengendalian internal nantinya menghasilkan kinerja yang efektif
dapat mencegah dan mengurangi tindakan dan mampu memberikan hasil yang baik
untuk pihak-pihak yang berkepentingan.

120
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun 2019 e-ISSN:
2614 – 1930

Hasil pengujian hipotesis ketiga ini cenderung akan berperilaku tidak etis dan
juga didukung oleh hasil penelitian yang berlaku curang untuk memaksimalkan
dilakukan Vani Adelin (2013), penegakan keuntungan untuk dirinya sendiri. Biasanya
peraturan dapat mencegah dan me- karyawan akan melakukan kecurangan
ngurangi tindakan kecurangan akuntansi, karena ketidakpuasaan atau kekecewaan
yang berarti bahwa semakin baik pene- dengan hasil atau kompensasi yang
gakan peraturan pada suatu perusahaan mereka terima atas apa yang telah di-
maka semakin rendah pula kecurangan kerjakan. Dengan kompensasi yang sesuai
akuntansi yang akan terjadi. diharapkan maka akan dapat memini-
malisasi tindakan kecurangan akuntansi.
Pengaruh Kesesuaian Kompensasi
terhadap Kecurangan Akuntansi Pengaruh Budaya Organisasi terhadap
Hasilpengujianhipotesiskeempatmeng Kecurangan Akuntansi
enaipengaruh kesesuaian kompen-sasi Hasilpengujianhipotesiskelima me-
terhadap kecurangan akuntansi me- ngenaipengaruh budaya organisasi ter-
nunjukkan nilai koefisien regresi -0,161 hadap kecurangan akuntansi menunjukkan
dengan nilai signifikansi uji t sebesar 0,042. nilai koefisien regresi -0,129 dengan nilai
Oleh karena itu, hipotesis keempat dalam signifikansi uji t sebesar 0,033. Oleh karena
penelitian ini diterima. Hal ini menunjukkan itu, hipotesis kelima dalam penelitian ini
bahwa kesesuaian kompensasi ber- diterima. Hal ini menunjukkan bahwa
pengaruh negatifsignifikan terhadap ke- budaya organisasi berpengaruh negatifsig-
curangan akuntansi. Jika kesesuaian nifikan terhadap kecurangan akuntansi. Jika
kompensasi semakin baik, maka budaya organisasi semakin baik, maka
kecurangan akuntansi semakin rendah. kecurangan akuntansi semakin rendah.
Hasil pengujian hipotesis keempat ini Hasil pengujian ini sependapat
sejalan dengan teori yang dinyatakan oleh dengan teori yang dinyatakan oleh
Edwin B. Flippo dalam Ryan Muhammad Moeljono dalam Zulkarnain (2013) bahwa
dan Ridwan (2017) dimana kompensasi budaya organisasi adalah sistem nilai-nilai
didefinisikan sebagai balas jasa yang adil yang diyakini semua anggota organisasi
dan layak diberikan kepada karyawan atas dan yang dipelajari, diterapkan serta dikem-
jasa-jasanya dalam mencapai tujuan bangkan secara berkesinambungan, ber-
organisasi. Sependapat dengan Edwin, fungsi sebagai sistem perekat, dan dapat
Andrew F. Sikula dalam Hasibuan dijadikan acuan berperilaku dalam organi-
(2003:119) mendefinisikan kompensasi sasi untuk mencapai tujuan organisasi yang
adalah segala sesuatu yang dianggap telah ditetapkan. Budaya organisasi meru-
sebagai balas jasa atau ekuivalen. pakan faktor yang paling kritis dalam
Kompensasi adalah salah satu hal yang organisasi.
penting bagi setiap karyawan yang bekerja Robbins dan Judge (2013) juga
dalam suatu perusahaan. Kecurangan menyatakan, jika budaya organisasi itu kuat
akuntansi dapat terjadi ketika seorang dan mendukung standar etika yang tinggi,
karyawan merasa tidak puas atas maka akan memiliki pengaruh yang sangat
kompensasi yang ia terima dari apa yang positif terhadap perilaku karyawannya.
telah mereka kerjakan. Tindakan ke- Sehingga ketika budaya organisasi tersebut
curangan akuntansi tersebut dilakukan baik, maka hal ini akan mendorong
karyawan semata-mata untuk memaksimal- karyawan untuk terhindar dari tindakan
kan keuntungan pribadi. kecurangan akuntansi.
Hasil pengujian hipotesis keempat ini Simpulan
juga didukung dengan hasil penelitian yang Penelitian ini bertujuan untuk mem-
dilakukan oleh Prekanida (2015) yang me- buktikan pengaruh dari variabel posisi
nyatakan bahwa kesesuaian kompensasi jabatan, keefektifan pengendalian internal,
juga merupakan faktor yang mempengaruhi penegakan peraturan, kesesuaian kompen-
terjadinya kecurangan akuntansi. Kompen- sasi dan budaya organisasi terhadap ke-
sasi merupakan hal yang berpengaruh curangan akuntansi yang terjadi di Bank
terhadap perilaku karyawan, seseorang Perkreditan Rakyat di Kabupaten Buleleng.

121
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun 2019 e-ISSN:
2614 – 1930

Berdasarkan hasil analisis data dan penelitian, maka untuk penelitian


pengujian hipotesis, maka dapat ditarik selanjutnya perlu ditambahkan metode
kesimpulan bahwa: wawancara langsung pada masing-
1. Posisi jabatan yang diperoleh oleh masing responden dalam upaya
seorang karyawan berpengaruh signifi- mengumpulkan data, sehingga dapat
kan positif terhadap kecurangan akun- menambah keabsahan data yang
tansi pada Bank Perkreditan Rakyat di diperoleh dan lebih mendalami bahan
Kabupaten Buleleng dengan nilai kajian penelitian.
koefisien regresi 0,124 dengan nilai 2. Penelitian ini hanya menggunakan 5
signifikansi uji t sebesar 0,037. variabel yaitu posisi jabatan, keefektifan
2. Keefektifan pengendalian internal yang pengendalian internal, penegakan pe-
diterapkan pada Bank Perkreditan raturan, kesesuaian kompensasi dan
Rakyat di Kabupaten Buleleng ber- budaya organisasi, hendaknya untuk
pengaruh signifikan negatif terhadap peneliti selanjutnya dapat menambahkan
kecurangan akuntansi pada Bank Per- beberapa variabel lagi sehingga faktor-
kreditan Rakyat di Kabupaten Buleleng faktor yang mempengaruhi terjadinya
dengan nilai koefisien regresi -0,130 kecurangan akuntansi akan lebih mudah
dengan nilai signifikansi uji t sebesar untuk ditemukan dan penelitian akan
0,026. menjadi lebih komprehensif.
3. Penegakan peraturan yang ada pada
Bank Perkreditan Rakyat di Kabupaten Bagi Bank Perkreditan Rakyat (BPR)
Buleleng berpengaruh signifikan negatif 1. Dalam kaitannya dengan pengaruh
terhadap kecurangan akuntansi pada posisi jabatan terhadap kecurangan
Bank Perkreditan Rakyat di Kabupaten akuntansi diharapkan dari pihak Bank
Buleleng dengan nilai koefisien regresi - Perkreditan Rakyat untuk lebih berhati-
0,140 dengan nilai signifikansi uji t hati dan selektif dalam memilih
sebesar 0,035. seseorang untuk ditempatkan pada
4. Kesesuaian kompensasi yang diperoleh posisi yang lebih tinggi agar nantinya
karyawan pada Bank Perkreditan Rakyat posisi yang didapatkan dan wewenang
di Kabupaten Buleleng berpengaruh yang dimilikinya dapat digunakan untuk
signifikan negatif terhadap kecurangan mencapai visi dan misi perusahaan.
akuntansi pada Bank Perkreditan Rakyat 2. Dalam meningkatkan pengawasan
di Kabupaten Buleleng dengan nilai pengendalian internal, penegakan pera-
koefisien regresi -0,161 dengan nilai turan, maka pimpinan Bank Perkreditan
signifikansi uji t sebesar 0,042. Rakyat dapat melakukan monitoring
5. Budaya organisasi yang diterapkan pada kerja terhadap karyawan pada waktu
Bank Perkreditan Rakyat di Kabupaten tertentu sehingga kinerja karyawan
Buleleng berpengaruh signifikan negatif dapat dipantau dengan efektif.
terhadap kecurangan akuntansi pada 3. Dalam kaitannya dengan kesesuaian
Bank Perkreditan Rakyat di Kabupaten kompensasi, maka Bank Perkreditan
Buleleng dengan nilai koefisien regresi - Rakyat hendaknya dapat meningkatkan
0,129 dengan nilai signifikansi uji t kesejahteraan karyawan melalui pem-
sebesar 0,033. berian bonus atau pemberian balas jasa
yang sesuai dengan kinerja karyawan
Saran sehingga tindakan kecurangan akuntansi
Berdasarkan hasil pembahasan pene- dapat dihindari di masa mendatang.
litian dan simpulan di atas, maka saran 4. Dalam kaitannya dengan budaya
yang dapat diberikan adalah sebagai organisasi, maka Bank Perkreditan
berikut. Rakyat hendaknya mampu menciptakan
suasana kerja yang nyaman, aman dan
Bagi Peneliti Selanjutnya kondusif sehingga karyawan dapat
1. Penelitian ini hanya menggunakan bekerja dengan baik dan tanpa tekanan.
kuesioner untuk mengumpulkan data

122
JIMAT (Jurnal Ilmiah Mahasiswa Akuntansi) Universitas Pendidikan Ganesha, Vol: 10 No: 2 Tahun 2019 e-ISSN:
2614 – 1930

DAFTAR PUSTAKA Sugiyono. 2013. Metode Penelitian


Kuantitatif dan Kualitatif dan R&D.
Balipost. 2018. “Kasus Korupsi PD BPR Bandung : Alfabeta.
Buleleng”. Tersedia pada
www.balipost.com/news (Diakses ----------- 2014. Metode Penelitian
pada tanggal 6 September 2018) Pendidikan Pendekatan Kuantitatif,
Kualitatif, dan R&D. Bandung :
Bali.Tribunnews. 2015. “Kolektor BPR di Alfabeta.
Buleleng Tilep Uang Nasabah”.
Tersedia pada Sumbayak, Jelfani Saragih. 2017.
www.bali.tribunnews.com (Diakses “Pengaruh Keadilan Organi-sasi,
pada tanggal 6 September 2018) Sistem Pengendalian Intern,
Komitmen Organisasi, serta Gaya
Damayanti, Dionisia Nadya Sri. 2016. Kepemimpinan Terhadap Kecurangan
“Pengaruh Pengendalian Internal (Fraud (Studi Empiris pada Kantor
serta Moralitas Individu dengan Cabang Utama Perusahaan Leasing
Kecurangan Akuntansi (Studi di Kota Pekanbaru)”. Tersedia pada
Eksperimen Pada Pegawai Bagian JOM Fekon, Vol. 4 No.1. (Diakses
Keuangan Dan Akuntansi Universitas pada tanggal 11 September 2018)
Negeri Yogyakarta)”. Tersedia pada
Skripsi Fakultas Ekonomi : Tarigan, Laurensia Br. 2016. “Pengaruh
Universitas Yogyakarta. (Diakses Moralitas Individu, Asimetri Informasi,
pada tanggal 10 September 2018) Efektivitas Pengendalian Internal
serta Ketaatan Aturan Akuntansi
Frilia, Figna. 2015. “Pengaruh Asimetri Terhadap Kecenderungan
Informasi, Sistem Pengendalian Kecurangan Akuntansi (Studi Pada
Intern, serta Kesesuaian Kompensasi BUMD Provinsi Riau)”. Tersedia pada
Terhadap Kecenderungan Kecurang- JOM Fekon Vol. 3 No. 1. (Diakses
an Akuntansi (Studi Empiris pada pada tanggal 11 September 2018)
Bank Swasta di Kota Pekanbaru)”.
Tersedia pada JOM Fekon, Vol. 2 No. Rizky, Muammar dan Fauziah Aida Fitri.
1. (Diakses pada tanggal 10 2017. “Pengaruh Keefektifan
September 2018) Pengendalian Internal, Ketaatan
Aturan Akuntansi, Penegakan Hukum,
Nurmayanti. 2018. “Otoritas Jasa Keuangan serta Perilaku Tidak Etis Terhadap
(OJK) Ungkap Kasus Penyelewengan Kecurangan Akuntansi”.Tersedia
Kredit di BPR KS Bali Agung pada JIMEKA Vol.2 No. 2. (Diakses
Sedana”.Tersedia pada pada tanggal 11 September 2018)
www.liputan6.com (Diakses pada 6
September 2018) Dewi, Chindy Kurnia Rahma. 2017.
“Pengaruh Pengendalian Internal,
Radhiah, Tia. 2016. “Pengaruh Efektivitas Kesesuaian Kompensasi, Ketaatan
Pengendalian Internal, Kesesuaian Aturan Akuntansi, serta Perilaku
Kompensasi serta Moralitas Individu Tidak Etis Terhadap Kecenderungan
Terhadap Kecenderungan Kecurang- Kecurangan Akuntansi (Studi Empiris
an Akuntansi (Studi Kasus Pada PT. pada SKPD Kabupaten Bengkalis)”.
POS Indonesia KCU Kota Pekan- JOM Fekon Vol. 4, No. 1. Tersedia
baru)”. Tersedia pada JOM Fekon pada media publication.com (Diakses
Vol.3 No.1. (Diakses pada tanggal 11 pada tanggal 10 September 2018)
September 2018)

123
Review of Accounting Studies (2019) 24:891–926
https://doi.org/10.1007/s11142-019-09508-w

A hidden risk of auditor industry specialization:


evidence from the financial crisis

Cory Cassell 1 & Emily Hunt 2 & Gans Narayanamoorthy 3 & Stephen P. Rowe 4

Published online: 8 August 2019


# Springer Science+Business Media, LLC, part of Springer Nature 2019

Abstract
We identify situations in which auditor industry specialization could be detrimental for
audit outcomes. We predict that during periods of heightened industry-specific risk,
specialist auditors from the affected industry could struggle to secure and allocate
sufficient resources to mitigate the heightened risk because they have client portfolios
concentrated in the affected industry. Using a measure of office-level industry
concentration/specialization (as opposed to a market-based measure), we find that
banking auditor industry specialization is associated with higher audit quality and more
timely audits during the period before the financial crisis. However, during the financial
crisis, banking industry specialization is associated with lower audit quality and less
timely audits. Collectively, our results suggest that auditor industry specialization can
be detrimental in certain circumstances and that audit firms and audit regulators should
consider whether the audit markets have become too specialized to handle the resource
allocation problems that crisis situations present.

Keywords Auditor industry expertise . Auditor industry specialization . Workload


compression . Resource constraints . Financial crisis

JEL classification M41 . M42 . L11 . L25 . G38

* Stephen P. Rowe
srowe@walton.uark.edu
Cory Cassell
ccassell@walton.uark.edu
Emily Hunt
ehunt@business.msstate.edu
Gans Narayanamoorthy
gnarayan@tulane.edu

Extended author information available on the last page of the article


892 C. Cassell et al.

1 Introduction

Extensive research in accounting suggests that auditors classified as industry specialists


provide audits of higher quality (e.g., Solomon and Shields 1999; Balsam et al. 2003;
Krishnan 2003; Krishnan 2005; Reichelt and Wang 2010), and investors, auditors, and
regulators consider auditor industry expertise an indicator of audit quality (Christensen
et al. 2016).1 However, inferences about the benefits of auditor industry specialization
are generally based on analyses that use large pooled samples, where average (net)
benefits are measured across time. In contrast, we use theoretical arguments and
evidence from the finance and management literatures to identify situations in which
auditor industry specialization could be detrimental for audit outcomes.
Evidence from the finance and management literatures highlights the benefits of
diversification when skills and resources are transferable across segments (e.g., Rumelt
1982; Palepu 1985; Pandya and Rao 1998), with one such benefit being the effective
allocation of resources during times of high uncertainty, risk, or both. Given that audit
professionals routinely work on clients from multiple industries, we argue that the audit
market is characterized as one with transferability of resources and skills (certainly more
so than a company with multiple divisions of diversified businesses like General Electric,
where there is significant variability across products and needed skills). The audit firm
structure of partners that routinely oversee audits in different industries also facilitates
such reassignment. Collectively, these arguments suggest that there are situations in which
audit office portfolio diversification (i.e., nonspecialization) could be beneficial.
Our study investigates the performance of specialist auditors during periods of
heightened risk. In stable environments, industry specialists should be able to apply
their unique experience and knowledge in ways that yield better audit outcomes.
However, during periods of heightened industry-specific risk, specialists from the
affected industry could struggle to secure and allocate sufficient resources to mitigate
the heightened risk in the affected industry because, by construction, they have client
portfolios concentrated in the affected industry (i.e., there is insufficient resource slack).
In contrast, nonspecialist auditors have client portfolios less concentrated in the affected
industry, suggesting that they should be better able to secure and allocate internal
resources to clients in the affected industry.
We test our predictions by investigating the effects of the recent financial crisis on
audits of clients from the banking industry. The financial crisis and the banking industry
provide an ideal setting for our investigation because: 1) the financial crisis affected
virtually all banks in the industry, 2) bank clients’ risk increased relative to other clients
during the financial crisis, 3) the banking industry represents a significant risk exposure
for many audit offices, 4) significant cross-sectional variation exists across audit offices
in their exposure to the banking industry, and 5) bank audits require specialized
knowledge, suggesting that they could benefit from auditor industry expertise.2 We
argue that the financial crisis increased the riskiness of banking clients and the
regulatory and market scrutiny faced by banks’ external auditors. This, in turn, caused
1
In 2015, the PCAOB cited auditor industry specialization on its list of 28 audit quality indicators (PCAOB
2015).
2
While the financial crisis of 2008 was a very significant event for banks, bank crises are not uncommon.
Previous crises include the savings and loan crisis (late 1980s and early 1990s), the inflation crisis (late 1970s
and early 1980s), and many others. See Bordo and Haubrich (2009) for a detailed history.
A hidden risk of auditor industry specialization: evidence from the... 893

a shift in auditor attention and resources toward bank clients that bore increased risk
(Doogar et al. 2015).3 The strain on office-level resources was likely particularly acute
for offices with a large proportion of clients from the banking industry (i.e., audit
offices that specialized in the audits of banks).
We perform our tests using proxies for audit quality (whether the financial state-
ments are materially misstated as revealed by a subsequent restatement) and audit
timeliness (the delay in issuing the audit opinion and whether the client files the annual
report after the filing deadline). Palmrose and Scholz (2004) argue that financial
statement restatements provide direct evidence of inferior audit quality and indicate
the auditor’s failure to enforce the correct application of GAAP. Audit opinion delay is
a measure of audit timeliness and has been associated with insufficient personnel
resources (Behn et al. 2006) and inexperienced personnel (Knechel and Payne 2001).
Late filing of annual reports has been associated with the subsequent release of bad
news and lower abnormal returns (Chambers and Penman 1984).
Auditor industry specialization is an unobservable construct in the absence of
proprietary audit firm data. Prior studies of auditor specialization generally proxy for
auditor industry specialization using a market-based benchmark (i.e., the extent to
which a given firm specializes in an industry relative to other audit firms in the market).
In contrast, similar to Krishnan (2003) and Krishnan (2005), we focus on a measure of
office-level industry concentration/specialization that does not rely on a market-based
comparison to construct our variable of interest. Specifically, we measure banking
auditor industry specialization as the sum of audit fees collected from bank clients
divided by the sum of audit fees collected from all clients, where both are measured
annually at the office level. Our within audit office measure captures the likelihood that
individual auditors have experience auditing banks and is closely tied to the concept of
auditor expertise. For example, in their Concept Release on Audit Quality Indicators,
the PCAOB states: “Experience with a particular industry helps an auditor understand
the industry’s operating practices, the critical audit and accounting issues confronting
companies in that industry, and the best ways to resolve those issues to further audit
quality” (PCAOB 2015, A-7). The resulting PCAOB audit quality indicator is con-
structed by summing the experience of engagement team members (partners, managers,
specialists, etc.) in the client’s industry. Thus, the PCAOB appears to view auditor
industry expertise as an absolute (non-relative) metric that originates at the individual
level and manifests at the audit team level. Consistent with this, Carcello and Nagy
(2004) argue that industry expertise depends on industry concentration at the office or
team level. Solomon and Shields (1999) define industry specialists as auditors whose
training and experience is largely in a particular industry (i.e., at the individual level),
and Wright and Wright (1997) find that auditors with industry experience (but who are
not designated as industry specialists) make superior risk assessments relative to
auditors without industry experience.
Our use of a nonmarket-based measure of banking auditor industry specialization
yields two important benefits in our setting. The first is theoretical—by construction,
market-based measures inhibit some small audit offices from being classified as industry

3
Conversations with bank auditors suggest that audit resources were significantly reallocated during the
financial crisis, with more resources being committed to bank clients. One manager we spoke with was
reassigned during the financial crisis to focus on a large banking client.
894 C. Cassell et al.

specialists, even though they have significant industry-specific expertise, and promote
some audit offices to industry specialist status, even though they have limited exposure to
a given industry. This contradicts both theoretical arguments and empirical evidence that
suggest that auditor expertise is an individual-, team-, or office-level construct.4 The
second is practical—our use of an office-based measure that does not depend on other
audit firms in the market allows us to capture aspects of both banking auditor industry
specialization and banking auditor industry concentration simultaneously.5
In our primary tests, we use a sample of banking clients from 2004 through 2009,
where the financial crisis period includes 2008 and 2009. We use a difference-in-
differences research design to alleviate concerns that our results are attributable to
unobserved crisis-related covariates. We find that, before the financial crisis, banking
auditor industry specialization is associated with higher audit quality (as indicated by a
lower incidence of misstatements) and more timely audits (as indicated by a shorter
audit opinion delay and a lower incidence of late filings). However, during the crisis,
the results indicate that banking auditor industry specialization is associated with lower
audit quality (as indicated by a higher incidence of misstatements) and less timely
audits (as indicated by a longer audit opinion delay). We do not find a significant
association between the incidence of late filings and banking auditor industry special-
ization during the financial crisis, but the results suggest that the benefits of speciali-
zation observed in the period before the financial crisis are no longer significant.
We perform two sets of tests to support our inference that the observed deterioration in
the benefits of auditor industry specialization is driven by problems associated with
securing and allocating resources to clients from the banking industry. First, we expect
that the resource allocation problem is more acute for banking auditor industry specialists
that have a number of poorly performing banks in their portfolio because the demand for
additional resources should be highest for such offices. Consistent with this, we find some
evidence suggesting that our results are driven by offices with banks that performed
relatively poorly during the financial crisis (here, bank performance is based on office-
level bank ROA). Second, we expect that the resource allocation problem could be
detrimental for clients from other industries that engage banking industry specialist
auditors because other-industry clients (OICs) provide the supply of resources available
for reallocation to the banking industry. The results provide some evidence suggesting that
the adverse implications of banking auditor industry specialization during the financial
crisis period extend to all clients within the audit office, suggesting an office-wide resource
allocation problem attributable to the concentration of clients within the affected industry.

4
This problem is particularly acute in industries that are concentrated geographically because auditor industry
specialization measures that rely on a market-based benchmark do not adequately consider the makeup of the
clientele in a particular city. Consider the expected number of specialist audit firms in a very large banking
center (e.g., New York City), compared to a smaller banking center (e.g., San Antonio). According to the
PCAOB’s audit quality indicator, New York City would be expected to have many more auditors that are
classified as specialists than San Antonio. However, market-based measures of specialization do not allow for
such variance in the distribution of specialists across cities because they essentially cap the number of auditors
that can be classified as specialists in a city where there would otherwise be many specialists and elevate
nonspecialists to specialist status in cities where there would otherwise be no or few specialists. We
acknowledge that these problems are likely far less acute in industries that are more geographically dispersed.
5
Although not directly related to the motivation and contribution of our study, Minutti-Meza (2013) and
Audousset-Coulier et al. (2016) provide additional evidence suggesting problems with market-based measures
of auditor industry specialization.
A hidden risk of auditor industry specialization: evidence from the... 895

We make several contributions to the literature. First, our study contributes to the
literature that investigates potential risks associated with auditor industry specialization
by providing the first evidence suggesting that auditor industry specialization can be
detrimental for audit quality in certain circumstances. Prior work has investigated the
impact of litigation risk on auditor’s willingness to specialize (Hogan and Jeter 1999)
and client concerns about information leakage to peer firms that use the same audit firm
(Aobdia 2015). Although the benefits of diversification are well known, the risks
associated with auditor industry specialization have been largely ignored by accounting
academics, practitioners, and regulators. Collectively, our results provide compelling
evidence suggesting that, although auditor industry specialization is associated with
improved audit outcomes during relatively stable periods, specialist auditors (and their
clients) subject themselves to significant risk during crises. Our results are particularly
important because they suggest that this risk manifests during periods of crisis—when
high quality and timely audits are most critical. We acknowledge, however, that our
analyses focus on a single industry and rely on a measure of auditor industry special-
ization that differs from the measure used in most prior work that investigates the
effects of auditor industry specialization.
Second, although investors, auditors and the PCAOB consider auditor industry
specialization as an indicator of audit quality (Christensen et al. 2016; PCAOB
2015), our results suggest that auditor industry diversification can facilitate the alloca-
tion of resources to areas of need and mitigate issues associated with time/budget
constraints. Our results suggest that these benefits accrue to clients in the industry most
affected by the crises and to other clients in the office. Thus, our evidence supports the
PCAOB’s assertion that budget constraints hinder auditors’ ability to adequately
supervise and review their engagements (Public Company Accounting Oversight
Board (PCAOB). 2010). We acknowledge that our results do not suggest that auditor
industry specialization is detrimental, on average. Rather, our results suggest that
auditor industry specialization can be detrimental in certain circumstances and that
audit firms and audit regulators should consider whether the audit market, audit firms,
or audit offices have become too specialized to handle the resource allocation problems
that crisis situations present.
The paper proceeds as follows. Section 2 discusses prior literature and develops our
hypothesis. Section 3 discusses our research methods, sample selection, and descriptive
statistics. Section 4 presents results from our main tests, additional analyses, and
robustness tests. Section 5 summarizes our findings and concludes.

2 Prior literature and hypothesis development

2.1 Auditor industry specialization

Specialist auditors are often associated with better quality because they have specialized
training and experience related to their industry (Solomon and Shields 1999). Consistent
with this, prior literature provides extensive evidence of better financial reporting quality
for clients of industry specialist auditors. For example, Balsam et al. (2003) and Krishnan
(2003) find that clients of auditors that are national industry specialists report smaller
abnormal accruals, and Reichelt and Wang (2010) find that audit quality is higher for audit
896 C. Cassell et al.

firms classified as city and national industry specialists. In a related study, Krishnan
(2005) finds that clients of specialist auditors report earnings that reflect more timely
recognition of bad news than clients of nonspecialist auditors.6 Auditors that specialize in
the banking industry have also been shown to provide better audit outcomes for their
banking clients. Specifically, Taylor (2010) finds that, for accounts highly specific to the
banking industry (i.e., loan loss reserves), bank specialist auditors were more confident in
their assessment of inherent risk than other auditors.

2.2 Specialization versus diversification

While auditor industry specialization is beneficial in terms of the auditor’s knowledge


and experience in a given industry, diversification is a well-known management and
investing strategy that has its own benefits. A diversified company is one that serves a
wide range of markets (Chandler Jr. 1962; Berry 1975; Aaker 1980; Andrews 1980;
Gluck 1985). Some prior research argues that companies diversify to improve firm
performance. Diversification is associated with improved firm performance because it
enables a company to better allocate resources and manage risks.7 Consistent with this,
studies have documented a positive association between firm diversification and future
performance (Pandya and Rao 1998; Rumelt 1982). In contrast, a number of studies by
financial economists document a negative association between diversification and firm
value. For example, Berger and Ofek (1995) find that diversified firms trade at a
discount of approximately 15% compared to focused competitors in the same
industry, and Lang and Stulz (1994) find a negative association between Tobin’s Q
(market value divided by the book value of assets) and diversification. Collectively,
prior work provides mixed evidence on the association between diversification and firm
performance and value. Thus, it is important to evaluate situations in which diversifi-
cation is likely to be beneficial and when it is likely to be detrimental.
Theoretically, the optimal (value maximizing) level of diversification should be such
that the economies of scope balance the diseconomies of scale (Rumelt 1982). When a
firm operates in a set of related businesses, it can exploit its core factors, leading to
economies of scale and scope, efficient resource allocation, and specialized technical
and managerial skill. Consistent with this, Palepu (1985) distinguishes between com-
panies with predominantly related diversification and companies with predominantly
unrelated diversification and finds that the former have significantly higher profitability
growth. Similarly, Desai and Jain (1999) find that increasing corporate focus by
spinning off unrelated (noncore) divisions generates approximately 50% higher returns
to shareholders over a three-year timeframe compared to spin-offs that do not refocus
the firm. Overall, prior research indicates that diversification can be beneficial when
resources can be transferred easily across areas should significant risks or opportunities
arise. More importantly, we argue that the audit market is characterized as one with
transferability of resources and skills because audit professionals can be assigned to
clients in different industries in response to changes in risk.
6
Similarly, Beardsley et al. (2018) find that misstatements are negatively associated with audit office industry
diversification, except for clients that cluster within a specific industry within the office. Beardsley et al. (2018)
do not investigate the effects of audit office industry diversification during times of industry specific crisis.
7
For example, a diversified company can transfer funds from a cash surplus unit to a cash deficit unit (without
taxes or transaction fees), thereby reducing the variability of operating cash flows (Bhide 1990).
A hidden risk of auditor industry specialization: evidence from the... 897

2.3 Audit risk and the allocation of resources

Auditors are required to respond to events that increase the likelihood that the financial
statements will be materially misstated and to conduct the audit in a manner that
reduces audit risk to an appropriately low level. Prior literature has examined auditor
response to various types of risk including the risk of litigation (Ponemon 1992; Pratt
and Stice 1994; Simunic and Stein 1996; Krishnan and Krishnan 1996), earnings
manipulation (Krishnan et al. 2013), and fraud (Payne and Ramsay 2005), among
others. Research specifically examining banks further supports the notion that auditors
respond to bank risks (Fields et al. 2004; Doogar et al. 2015).
Auditors respond to changes in risk by adjusting the nature, timing, and extent of
audit procedures.8 For example, research has demonstrated that auditors allocate more
and better resources (i.e. personnel hours) to clients with higher risk (Bell et al. 2008).
Other research has documented a range of responses to increases in risk including
adjusting the audit plan and audit fees (Pratt and Stice 1994; Simunic and Stein 1996),
increasing the level of professional skepticism (Payne and Ramsay 2005), increasing
the issuance of modified opinions (Krishnan and Krishnan 1996), and resigning from
the client (Krishnan et al. 2013). Under normal circumstances, auditors can respond to
changes in risk by reallocating resources and making other adjustments. However, an
exogenous shock can hinder auditors’ ability to secure and allocate sufficient resources
needed to mitigate the heightened risk in the client portfolio. We argue that the financial
crisis of 2008 was such a shock that likely contributed to significant resource allocation
issues for some audit offices.

2.4 Audit risk in the banking industry and the financial crisis

In 2008, the deterioration of the economy increased risk across the board as many large
financial institutions failed and many others were acquired or recapitalized by the
federal government. On March 14, 2008, the Federal Reserve Bank of New York
and JPMorgan Chase announced their emergency cash bailout of Bear Stearns follow-
ing a significant deterioration of the firm’s liquidity.9 On July 11, 2008, IndyMac Bank
failed, marking one of the largest bank failures in American history, and the Federal
Deposit Insurance Corporation (FDIC) was named the conservator.10 Beginning in
October 2008 and continuing through 2009, the federal government invested billions of
dollars to bail out Fannie Mae, Freddie Mac, AIG, Citigroup, Bank of America,
JPMorgan Chase, Morgan Stanley, Goldman Sachs, Merrill Lynch, Regions Financial
Corporation, and hundreds of other financial institutions through its Troubled Asset
Relief Program (TARP).11
During the financial crisis, banks were under significant scrutiny from regulators and
the general public who wanted to understand the role they played in the subprime
mortgage crisis. Similarly, bank auditors were under intense pressure to explain how
they could have missed the warning signs (Bajaj and Creswell 2008). Consistent with
8
Public Company Accounting Oversight Board, Auditing Standard No. 8 and Auditing Standard No. 13.
9
“Bear Stearns Big Bailout.” Bloomberg, 2008. Matthew Goldstein.
10
“FDIC Establishes IndyMac Federal Bank, FSB as Successor to IndyMac Bank, F.S.B., Pasadena,
California.” FDIC Press Release, 2008.
11
“Bailed Out Banks.” CNN Money Special Report.
898 C. Cassell et al.

arguments in the preceding section, there is evidence suggesting that bank auditors
responded to the financial crisis in accordance with current auditing standards by
adjusting audit plans in response to the heightened risk. For example, Doogar et al.
(2015) use audit fees as a proxy for auditor attention and find that auditor attention
during the lead-up to the financial crisis shifted in line with varying sources of audit
risk.12

2.5 Auditor workload compression

The PCAOB has expressed concern about the effects of time constraints on auditors’
ability to adequately supervise and review the audit (Public Company Accounting
Oversight Board (PCAOB). 2010), suggesting that audit quality is diminished when
auditors are busy. A number of studies provide evidence consistent with this concern.
Bills, Swanquist, and Whited (Bills et al. 2015) document a decline in audit quality
consistent with resource constraints when audit firms experience substantial growth at
the office level. Coram et al. (2004) find that auditors engage in quality reduction acts
as a strategic response to time pressure. Lopez and Peters (2012) find that abnormal
accruals are larger and companies are more likely to meet or beat earnings benchmarks
for audits completed during the audit busy season.

2.6 Hypothesis development

Given that: 1) audit resources can be (relatively easily) transferred across clients within
an audit office (see Section 2.2), 2) audit firms allocate resources according to client
risks (see Section 2.3), 3) audit risks in the banking industry increased significantly
during the financial crisis (see Section 2.4), and 4) audit quality is negatively associated
with time pressure and workload compression (see Section 2.5)—we posit that the
demand for audit resources allocated to banking clients increased significantly during
the financial crisis. Presumably, additional resources would be secured from clients that
operate outside of the most affected industry (i.e., OICs). Thus, at the audit office level,
our prediction has two important implications: 1) the demand for additional resources is
higher in audit offices with more banking clients, and 2) the supply of available
resources is lower in audit offices with more banking clients. This, in turn, suggests
that audit offices with greater banking industry specialization could both have greater
demand for additional resources and the smallest pool of available resources from
which to draw.13 Stated differently, during periods of heightened industry-specific risks,
auditors who specialize in the affected industry are particularly vulnerable to the

12
Michael Young, a lawyer at Willkie Farr & Gallagher who specializes in cases involving accounting
irregularities, states: “Auditors have actually been pretty tough during the crisis in forcing companies to justify
their valuation methods, a move which has resulted in many banks having to write down the value of their
assets.” See “Role of Auditor in Crisis Gets Look.” Wall Street Journal, 2010. Michael Rapaport.
13
A number of factors provide tension for our hypothesis. First, offices (or firms) with a large number of
banking clients may have been better equipped to respond to the implications of the crisis (e.g., with training,
distribution of best practices, etc.) such that banking specialization became more beneficial for audit outcomes
in the crisis period despite constrained personnel resources. Second, to the extent that crisis-related increases in
risk were gradual or anticipatable, auditors could likely secure additional resources to mitigate the heightened
risk. Finally, audit offices could request and secure resources from affiliated offices (i.e., within the operating
region of the firm). We discuss tests that address this possibility in Section 4.3.1.
A hidden risk of auditor industry specialization: evidence from the... 899

adverse implications of budget constraints and workload compression. The preceding


discussion forms the basis for our hypothesis, stated in the alternative form.

H1: The benefits of banking auditor industry specialization for bank client audit
outcomes are mitigated during the financial crisis, compared to the pre-crisis
period.

3 Research methods, sample selection, and descriptive statistics

3.1 Research methods

We perform tests using three observable audit outcomes that capture the quality
(misstatements) and timeliness (audit opinion delay and late filings) of the audit. We
use severe (i.e., so-called big R) financial statement misstatements (MISSTATE) to
proxy for low quality audits. These restatements are reported through Form 8-K filings
with the SEC.14 Misstatements represent material errors or omissions in previously
issued financial statements (as revealed through subsequent restatement announce-
ments) and have been used extensively in prior research as a measure of audit and
financial reporting quality (Becker et al. 1998; Francis et al. 1999; Liu et al. 2009).
We use audit opinion delay (DELAY) and late filings (LATE_FILE) to proxy for
audit timeliness. Delayed financial reports impair relevancy by reducing the timeliness
of the information provided to investors.15 Longer delays reduce the information
content of the financial statements as evidenced by less pronounced investor reactions
(Givoly and Palmon 1982; Atiase et al. 1989). Delayed filings also allow subsets of
investors to acquire private information, which compromises the ideal of equal access
to information among investors.16
Ultimately, audit opinion delays can result in companies failing to file their annual
financial statements by the deadline imposed by the SEC. The SEC requires that annual
reports containing financial statements be filed within: 1) 90 days after the fiscal year-
end for non-accelerated filers, 2) 75 days after the fiscal year-end for accelerated filers,
and 3) 60 days after the fiscal year-end for large accelerated filers. Companies that fail
to file an annual report in a timely manner are required to inform investors via form
12b-25. Feldman et al. (2006) find that the filing of form 12b-25 elicits a negative

14
DeAngelo (1981) defines audit quality as the probability of detecting material errors and misstatements in
the client’s financial statements and reporting the detected errors and misstatements. Misstatements represent a
verifiable occurrence of poor financial reporting quality (DeFond 2010) and are a major focus of regulator and
investor concerns about the quality of audits (Myers et al. 2003). Similarly, Palmrose and Scholz (2004) argue
that financial statement restatements provide direct evidence of inferior audit quality and indicate the auditor’s
failure to enforce the appropriate application of GAAP.
15
According to the Financial Accounting Standards Board (FASB) Statement of Financial Accounting
Concepts No. 2, “Timeliness alone cannot make information relevant, but a lack of information timeliness
can rob information of relevance it might otherwise have had.”
16
In 2002, the Securities and Exchange Commission (SEC) expressed concerns about equal access to
information among investors and required CEOs to personally vouch for the timeliness and fairness of public
information disclosures (see https://www.sec.gov/news/press/2002-88.htm). The Dodd Frank Act further
increases enforcement of equal access to information by imposing monetary penalties and sanctions for
insider trading.
900 C. Cassell et al.

market reaction, suggesting that there is a significant cost for failing to meet the filing
deadline.
Prior literature identifies a number of factors associated with budget pressures and
workload compression that contribute to delayed audit opinions and late filings. For
example, Behn et al. (2006) find that resource constraints (related to both the client and
the auditor) hinder significant reduction in audit opinion delay. Other studies provide
evidence of an association between audit opinion delay and a sudden increase in client
size (Henderson and Kaplan 2000), the correction of previously reported interim
earnings (Kinney and McDaniel 1993), the risk of the underlying audit (Ashton et al.
1989), the use of inexperienced audit staff (Knechel and Payne 2001), and the
discovery of internal control weaknesses (Ettredge et al. 2006).
We use an interaction model (difference-in-differences design) to test H1.17 We
regress our three proxies for audit outcomes (MISSTATE, DELAY, and LATE_FILE) on
the interaction of banking auditor industry specialization and an indicator for the
financial crisis period along with main effects and other controls. We estimate the
following model using ordinary least squares (OLS)18:

DV it ¼ α0 þ α1 BANK SPEC it þ α2 CRISIS it þ α3 BANK SPEC it *CRISIS it


þ α4 LEVERAGEit þ α5 CLIENTSIZEit þ α6 GCOit þ α7 BIG4it

þ α8 LOSS it þ α9 ROAit þ α10 TENUREit þ α11 INFLUENCE it


þ α12 INTANGIBLES it þ α13 FOR OPS it þ α14 L3 ASSETS it þ α15 BUSY it

þ α16 EXCHANGEit þ α17 BTM it þ α18 ACQUISITION it þ α19 AFILERit


þ α20 AFILER LARGEit þ ε ð1Þ

DV is MISSTATE, DELAY, or LATE_FILE, depending on the specification. MISSTATE


is an indicator variable set equal to one if the company has a financial statement
misstatement reported through a Form 8-K filing with the SEC (as reported by Audit
Analytics), and zero otherwise. DELAY is the number of days between the fiscal year-
end and the audit opinion date.19 LATE_FILE is an indicator variable set equal to one if
the company files its financial report after the SEC deadline, and zero otherwise.20
BANK_SPEC is the sum of audit fees collected from banking clients, divided by the
sum of audit fees collected from all clients, where both inputs are measured annually at
the office level (audit fees and audit office are provided by Audit Analytics). Following
17
In this quasi-experimental design, we use the financial crisis as a treatment condition, whereby we examine
the difference in the effects of banking auditor industry specialization during the financial crisis and the period
before the crisis.
18
Following Shipman et al. (2017), we use a linear probability model to estimate Equation (1) because this
allows us to preserve sample observations, compare average treatment effects, and interpret the interaction
term as a difference-in-differences test. Statistical inferences are similar for both LATE_FILE and MISSTATE
when we use a probit model and estimate the interaction effect following Norton et al. (2004).
19
To facilitate the interpretation of coefficients, we tabulate results for raw (untransformed) DELAY. However,
because raw DELAY does not meet the normal distribution assumption required by OLS (Knechel and Sharma
2012), we also perform tests using the natural log of DELAY and using a negative binomial model. Inferences
from the (untabulated) results are similar. All untabulated analyses are available from the authors upon request.
20
LATE_FILE is constructed using the Audit Analytics non-timely filer database.
A hidden risk of auditor industry specialization: evidence from the... 901

Cull and Peria (2013), CRISIS is an indicator variable set equal to one for company-
years 2008 and 2009, and zero otherwise.21
Equation (1) includes controls for variables that have been shown to affect audit
outcomes in previous work (Ettredge et al. 2006; Lennox and Li 2014). LEVERAGE is
equal to total liabilities divided by total assets. CLIENTSIZE is the natural logarithm of
total assets in millions. GCO is an indicator variable set equal to one if the company
receives a going-concern opinion, and zero otherwise. BIG4 is an indicator variable set
equal to one if the company engages a Big 4 auditor, and zero otherwise. LOSS is an
indicator variable set equal to one if net income is less than zero, and zero otherwise.
ROA is equal to income before extraordinary items divided by total assets. TENURE is
an indicator variable set equal to one if the current auditor’s tenure is more than four
years, and zero otherwise. INFLUENCE is equal to company audit fees divided by the
sum of audit fees collected from all clients in the office. INTANGIBLES is equal to
intangible assets divided by total assets. FOR_OPS is an indicator variable set equal to
one if the company reports a currency translation adjustment after net income to arrive
at total comprehensive income, and zero otherwise. L3_ASSETS is an indicator variable
set equal to one if the company has level 3 fair value assets (which would have been
particularly difficult to value during the financial crisis), and zero otherwise. BUSY is an
indicator variable set equal to one if the company’s fiscal year-end is in November,
December or January, and zero otherwise. EXCHANGE is an indicator variable set
equal to one if the company trades on a major exchange (NYSE, American,
NASDAQ), and zero otherwise. BTM is the book value of equity divided by the market
value of equity. ACQUISITION is an indicator variable set equal to one if the company
reports an acquisition, and zero otherwise. AFILER is an indicator variable set equal to
one if the company is an accelerated filer, and zero otherwise. AFILER_LARGE is an
indicator variable set equal to one if the company is a large accelerated filer, and zero
otherwise.22 Lastly we control for year and MSA fixed effects in all tests, and we
cluster standard errors by company.23
In Eq. (1), the coefficient on BANK_SPEC (α1) captures the effect of banking
auditor industry specialization on audit outcomes during the period before the financial
crisis. The coefficient on CRISIS (α2) captures the effect of the crisis on audit outcomes
when BANK_SPEC equals zero (i.e., audit offices with no banking clients). The
coefficient on the interaction between BANK_SPEC and CRISIS (α3) captures the
incremental effect of banking auditor industry specialization on audit outcomes during
the financial crisis relative to the pre-crisis period. H1 predicts a positive and significant
coefficient on the interaction term, which would indicate that the benefits of banking
auditor industry specialization (if any) are mitigated or reversed during the financial
crisis.

21
We designate 2008 and 2009 as crisis years; however, there is substantial divergence of opinion as to what
the financial crisis years were. Cullen et al. (2018) use bankruptcies to indicate that the crisis began in late
2007, a perspective shared by the National Bureau of Economic Research. (http://www.nber.org/cycles.html).
Other bank auditing research (Doogar et al. 2015) highlights 2007 as the “lead-up” to the financial crisis.
Regardless of differences in opinions about when the financial crisis began, most research supports 2008 and
2009 as crisis periods.
22
All continuous control variables are winsorized at the first and 99th percentiles.
23
The (untabulated) results are similar if we exclude year and MSA fixed effects or cluster standard errors at
the audit-office level.
902 C. Cassell et al.

3.2 Sample selection and descriptive statistics

The sample construction process begins with all company-year observations from 2004
through 2009 with necessary data in Compustat and Audit Analytics to construct the
variables in Eq. (1). The sample is restricted to clients in the banking industry, classified
as SIC codes 60 through 62 which include depository institutions, nondepository
institutions, and security and commodity brokers.24 We also restrict the sample to
domestic banks with a domestic auditor. Finally, we exclude audit firms with fewer
than 10 public clients to ensure that our measure of banking auditor industry special-
ization is meaningful.25 The final sample consists of 3,199 bank company-year
observations.
We present comparative descriptive statistics (financial crisis and pre-crisis periods)
in Table 1, Panel A, and the distribution of bank specialization by audit firm in Table 1,
Panel B. By construction, the bulk of the sample falls in the pre-crisis period, with
approximately 30% of the observations falling in the crisis period. The incidence of
misstatements (MISSTATE) is very low in both periods, with only 1.2 (3.7) percent of
banks having a misstatement during the financial crisis (pre-crisis) period. As a result,
for analyses where the dependent variable is MISSTATE, our inferences are based on
small samples (e.g., there are only 12 misstatements in the crisis period).26 The average
audit opinion delay (DELAY) is approximately 73 and 65 days after the fiscal year-end
during the crisis and pre-crisis periods, respectively. The incidence of late financial
statement filings (LATE_FILE) is relatively stable across the two periods, with approx-
imately 7% to 8% of banks failing to file by the deadline. The average of BANK_SPEC
is 58.5% in the crisis period and 52.8% in the pre-crisis period. Notably, the sample
exhibits significant variance in BANK_SPEC in both periods (with ranges from less
than 1% to 100% and large standard deviations), suggesting that there is significant
variance in the extent of banking auditor industry specialization in the audit offices
included in the sample. Finally, as expected, average financial performance (ROA) is
worse and the incidence of losses (LOSS) is higher during the financial crisis.27
In Table 1, Panel B, we present the distribution of bank specialization by audit firm
to provide insight into our bank specialization measure. Of note, 26 of the 56 audit
firms have at least one office where 100% of the public clients are banks (this includes
each of the Big 4 firms). As would be expected, Big 4 audit offices are typically larger

24
Our sample size is larger than that of bank studies that restrict their analyses to bank holding companies
(i.e., those that use regulatory reporting data sets). A limitation of the expanded sample is that we cannot
control for certain bank characteristics (e.g., securitizations) that are available in the regulatory reporting data
sets.
25
We calculate BANK_SPEC using the full Audit Analytics audit fees dataset (before merging with
Compustat). As discussed in Section 4.3.5, we find similar results if we calculate BANK_SPEC using the
final sample.
26
In (untabulated) analyses, we perform tests using a less restrictive measure of MISSTATE that captures both
big R and little r misstatements. By definition, little r restatements are less severe, as they are not required to be
reported through an 8-K filing and do not require revision or withdrawal of the audit opinion (Tan and Young
2015). In these tests, we find insignificant coefficients on both BANK_SPEC and the interaction between
BANK_SPEC and CRISIS.
27
The percentage of banks audited by Big 4 firms was 45% during the pre-crisis period (consistent with
previous banking literature; Doogar et al. 2015) and decreased to 35% during the crisis. Untabulated analysis
of Big 4 versus non-Big 4 clients from 2007 to 2009 reveals that, while a similar proportion of non-Big 4
banks dropped out of the sample, non-Big 4 firms attracted a larger proportion of banks that switched auditors.
A hidden risk of auditor industry specialization: evidence from the... 903

than other audit offices, which leads to fewer Big 4 offices with a high proportion of
bank audits compared to non-Big 4 offices. Overall, the results in Table 1, Panel B,
indicate that there is significant variance in our bank specialization measure—both
within audit firms across offices, and across audit firms.28
We present Pearson (below the diagonal) and Spearman (above the diagonal)
correlations in Table 2. Across the pre-crisis and crisis periods combined, BANK_SPEC
is significantly negatively associated with MISSTATE and negatively but insignificantly
associated with DELAY and LATE_FILE, suggesting that banking auditor industry
specialization is associated with better audit outcomes, on average (i.e., across the
sample period). Other control variables are associated with our outcome variables
consistent with expectations. For example, large banks, banks with higher ROA, and
banks with longer TENURE are associated with shorter DELAY and are less likely to
LATE_FILE.

4 Results

4.1 Tests of H1

Results from the estimation of Eq. (1), used to test H1, are presented in Table 3 and
depicted in Fig. 1. In Table 3, the dependent variable is MISSTATE, DELAY, and
LATE_FILE in Columns (1), (2), and (3), respectively. In all columns, the coefficient
on BANK_SPEC is negative and statistically significant (p < 0.01 or p < 0.10, depend-
ing on the specification). This indicates that, during the pre-crisis period, audit out-
comes improve as the extent of banking auditor industry specialization increases. As
discussed previously, this result is consistent with a large body of research on the effects
of auditor industry specialization.
Importantly, the coefficient on the interaction term (BANK_SPEC x CRISIS) is
positive and significant in each column (p < 0.01). Thus, consistent with H1, the
benefits of banking industry specialization (observed during the pre-crisis period) are
mitigated during the crisis. Note that H1 predicts only that the benefits of banking
industry specialization will be mitigated during the crisis—that is, we do not develop a
formal hypothesis about the joint test because doing so would require predictions about
the magnitude of the effect of the financial crisis on the relation between audit
outcomes and auditor industry specialization. Nevertheless, we consider the joint effect
in exploratory tests. The joint tests indicate that the association between BANK_SPEC
and two of the three outcome variables (MISSTATE and DELAY) is positive and
significant (p < 0.05 and p < 0.01, respectively). The joint test is insignificant in
Column (3), where the dependent variable is LATE_FILE. Thus, the results suggest
that, during the financial crisis, banking industry specialization was detrimental for
audit outcomes (in terms of the incidence of misstatements and audit opinion delays) or
had no statistically detectable benefit (in terms of late filings).
In Fig. 1, we present predicted marginal means from the estimation of Eq. (1) with
our continuous BANK_SPEC variable replaced with a dichotomous high (low) bank

28
As discussed in Section 4.3.5, we find similar results if we drop offices with fewer than 10 public clients or
more than 75 public clients and when we drop the five largest banking cities.
904

Table 1 Summary statistics

Panel A: Financial Crisis and Pre-Crisis Period Observations


Variable Financial Crisis Pre-Crisis Difference
N Mean S.D. 25% Median 75% N Mean S.D. 25% Median 75%
MISSTATE 969 0.012 0.111 0.000 0.000 0.000 2230 0.037 0.189 0.000 0.000 0.000 (0.025)***
DELAY 969 72.643 16.287 61.000 71.000 82.000 2230 64.856 19.128 57.000 67.000 74.000 7.787***
LATE_FILE 969 0.084 0.277 0.000 0.000 0.000 2230 0.072 0.258 0.000 0.000 0.000 0.012
BANK_SPEC 969 0.585 0.373 0.187 0.655 1.000 2230 0.528 0.381 0.146 0.481 1.000 0.057***
LEVERAGE 969 0.868 0.152 0.881 0.905 0.923 2230 0.872 0.159 0.887 0.908 0.923 (0.004)
CLIENTSIZE 969 7.379 1.716 6.388 7.170 8.133 2230 7.193 1.806 6.207 6.941 8.021 0.187***
GCO 969 0.046 0.211 0.000 0.000 0.000 2230 0.009 0.092 0.000 0.000 0.000 0.038***
BIG4 969 0.349 0.477 0.000 0.000 1.000 2230 0.447 0.497 0.000 0.000 1.000 (0.098)***
LOSS 969 0.223 0.416 0.000 0.000 0.000 2230 0.048 0.213 0.000 0.000 0.000 0.175***
ROA 969 −0.013 0.128 −0.010 0.002 0.007 2230 0.005 0.105 0.006 0.009 0.012 (0.018)***
TENURE 969 0.286 0.452 0.000 0.000 1.000 2230 0.283 0.450 0.000 0.000 1.000 0.003
INFLUENCE 969 0.158 0.216 0.039 0.082 0.170 2230 0.137 0.192 0.024 0.067 0.164 0.021***
INTANGIBLES 969 0.025 0.073 0.000 0.006 0.025 2230 0.025 0.061 0.000 0.008 0.026 (0.001)
FOR_OPS 969 0.520 0.500 0.000 1.000 1.000 2230 0.548 0.498 0.000 1.000 1.000 (0.028)
L3_ASSETS 969 0.768 0.422 1.000 1.000 1.000 2230 0.998 0.047 1.000 1.000 1.000 (0.230)***
BUSY 969 0.897 0.304 1.000 1.000 1.000 2230 0.879 0.326 1.000 1.000 1.000 0.017
EXCHANGE 969 0.807 0.395 1.000 1.000 1.000 2230 0.765 0.424 1.000 1.000 1.000 0.042***
BTM 969 1.444 1.136 0.812 1.195 1.915 2230 0.616 0.534 0.444 0.584 0.777 0.828***
ACQUISITION 969 0.547 0.498 0.000 1.000 1.000 2230 0.588 0.492 0.000 1.000 1.000 (0.041)**
AFILER 969 0.353 0.478 0.000 0.000 1.000 2230 0.369 0.483 0.000 0.000 1.000 (0.016)
AFILER_LARGE 969 0.270 0.444 0.000 0.000 1.000 2230 0.252 0.434 0.000 0.000 1.000 0.018
C. Cassell et al.
Table 1 (continued)
Panel B: Bank Specialization by Audit Firm
Auditor name Number of Offices Mean Number Bank Fees as Percentage of Total Fees
with Public Bank Clients of Public
Clients
All Bank 1% 25% Median 75% 99%
KPMG LLP 52 41 8 0.01 0.10 0.22 0.42 1.00
Deloitte & Touche LLP 36 62 9 0.01 0.04 0.08 0.31 1.00
Ernst & Young LLP 31 76 5 0.02 0.06 0.11 0.19 1.00
PricewaterhouseCoopers LLP 28 68 6 0.01 0.06 0.19 0.32 1.00
McGladrey & Pullen LLP 21 5 3 0.10 0.41 0.73 1.00 1.00
Grant Thornton LLP 16 19 7 0.02 0.15 0.24 0.48 1.00
BKD LLP 15 7 7 0.06 0.86 1.00 1.00 1.00
Crowe Horwath LLP (Post 2008) 14 10 9 0.25 0.78 1.00 1.00 1.00
Crowe Chizek & Company LLP (Pre- 11 10 9 0.48 0.87 1.00 1.00 1.00
2008)
BDO Seidman LLP 9 9 3 0.02 0.21 0.34 0.41 0.84
A hidden risk of auditor industry specialization: evidence from the...

Beard Miller Company LLP 9 8 7 0.37 0.44 1.00 1.00 1.00


Moss Adams LLP 7 7 5 0.09 0.78 1.00 1.00 1.00
ParenteBeard LLC 6 8 7 0.38 0.74 0.87 1.00 1.00
Dixon Hughes PLLC 6 7 6 0.59 0.78 0.90 1.00 1.00
Elliott Davis LLC/PLLC 5 12 10 0.74 0.81 0.86 1.00 1.00
Plante & Moran PLLC 5 8 6 0.48 0.76 0.92 1.00 1.00
Hacker Johnson & Smith PA 4 4 4 0.63 0.73 1.00 1.00 1.00
Cherry Bekaert & Holland LLP 4 2 1 0.19 1.00 1.00 1.00 1.00
Virchow Krause & Company LLP 3 16 2 0.09 0.20 0.23 1.00 1.00
Mauldin & Jenkins LLC 3 7 7 1.00 1.00 1.00 1.00 1.00
905
Table 1 (continued)
906

Vavrinek Trine Day & Company LLP 3 6 6 1.00 1.00 1.00 1.00 1.00
UHY LLP 3 5 1 0.08 0.13 0.28 0.30 0.45
Sherb & Co LLP 2 31 2 0.13 0.16 0.19 0.21 0.22
Marcum & Kliegman LLP 2 28 2 0.03 0.05 0.08 0.11 0.15
Bagell Josephs Levine & Co LLC 2 23 3 0.25 0.25 0.27 0.41 0.41
Perry-Smith LLP 2 13 8 0.40 0.67 0.71 0.83 0.86
SR Snodgrass PC/AC 2 13 13 1.00 1.00 1.00 1.00 1.00
Burr Pilger Mayer Inc 2 11 1 0.05 0.08 0.09 0.14 0.15
Stonefield Josephson Inc 2 9 1 0.03 0.03 0.03 0.09 0.14
Lazar Levine & Felix LLP 2 8 1 0.10 0.14 0.59 1.00 1.00
Baker Tilly Virchow Krause LLP 2 2 1 0.35 0.35 0.68 1.00 1.00
Eisner LLP 1 49 8 0.10 0.12 0.13 0.13 0.16
Goldstein Golub Kessler LLP (GGK) 1 33 1 0.02 0.02 0.02 0.02 0.02
Hansen Barnett & Maxwell PC 1 28 1 0.15 0.15 0.15 0.16 0.18
Beckstead & Watts LLP 1 27 2 0.26 0.26 0.26 0.26 0.26
Caturano & Company Inc./PC 1 26 2 0.06 0.06 0.06 0.06 0.06
Vitale Caturano & Company Ltd 1 24 2 0.07 0.07 0.07 0.09 0.09
Rothstein Kass & Company PC 1 22 3 0.14 0.14 0.26 0.29 0.29
Marcum LLP 1 22 2 0.11 0.11 0.11 0.11 0.11
Yount Hyde & Barbour PC 1 21 21 0.96 0.96 1.00 1.00 1.00
Peterson Sullivan LLP/PLLC 1 21 1 0.03 0.03 0.05 0.07 0.07
De Joya Griffith & Company LLC 1 19 1 0.06 0.06 0.06 0.06 0.06
Wolf & Company PC (MA) 1 18 12 0.61 0.63 0.74 0.82 0.86
Amper Politziner & Mattia LLP 1 18 1 0.01 0.01 0.01 0.02 0.02
Paritz & Company PA 1 17 1 0.07 0.07 0.07 0.10 0.10
C. Cassell et al.
Table 1 (continued)
Squar Milner Peterson Miranda & 1 17 1 0.17 0.17 0.17 0.17 0.17
Williams
Larry O’Donnell & Co CPA PC 1 17 2 0.02 0.06 0.12 0.12 0.20
Stegman & Company PA 1 16 10 0.62 0.63 0.64 0.71 0.78
Vitale Caturano & Co PC 1 15 1 0.07 0.07 0.07 0.07 0.07
KBA Group LLP 1 15 1 0.06 0.06 0.08 0.10 0.10
Porter Keadle Moore LLC 1 12 10 0.90 0.90 0.90 0.92 1.00
Miller Ellin & Co LLP 1 12 2 0.07 0.07 0.14 0.21 0.21
Weiser LLP 1 11 2 0.40 0.40 0.40 0.40 0.40
Demetrius & Company LLC 1 11 2 0.07 0.07 0.09 0.12 0.12
PMB Helin Donovan LLP 1 5 1 0.14 0.14 0.29 0.44 0.44
Mayer Hoffman McCann PC 1 1 1 1.00 1.00 1.00 1.00 1.00

Table 1 Panel B presents bank specialization descriptive statistics summarized by audit firm. The mean number of public clients (all and bank) is the average number of public clients in
each office for that firm over the sample period, based on all observations contained in Audit Analytics database. Bank fees as a percentage of total fees represents the distribution of our
primary variable of interest (BANK_SPEC) for each audit firm.
A hidden risk of auditor industry specialization: evidence from the...

Variable Definitions:
MISSTATE = Indicator variable set equal to one if the company has a financial statement misstatement reported through a Form 8-K filing with the SEC (as reported by Audit
Analytics), and zero otherwise.
DELAY = The number of days between the fiscal year-end and the audit opinion date.
LATE_FILE = Indicator variable set equal to one if the company files its financial report after the SEC deadline, and zero otherwise.
BANK_SPEC = The sum of audit fees collected from banking clients, divided by the sum of audit fees collected from all clients, where both inputs are measured annually at the office
level.
CRISIS = Indicator variable set equal to one for company-years 2008 and 2009, and zero otherwise.
LEVERAGE = Total liabilities divided by total assets.
CLIENTSIZE = The natural logarithm of total assets in millions.
GCO = Indicator variable set equal to one if the company receives a going concern opinion, and zero otherwise.
907
908

BIG4 = Indicator variable set equal to one if the company engages a Big4 auditor, and zero otherwise.
LOSS = Indicator variable set equal to one if net income is less than zero, and zero otherwise.
ROA = Income before extraordinary items divided by total assets.
TENURE = Indicator variable set equal to one if the current auditor’s tenure is more than four years, and zero otherwise.
INFLUENCE = Company audit fees divided by the sum of audit fees collected from all clients in the office.
INTANGIBLES = Intangible assets divided by total assets.
FOR_OPS = Indicator variable set equal to one if the company reports a currency translation adjustment after net income to arrive at total comprehensive income, and zero otherwise.
L3_ASSETS = Indicator variable set equal to one if the company has level 3 fair value assets, and zero otherwise.
BUSY = Indicator variable set equal to one if the company’s fiscal year-end is in November, December or January, and zero otherwise.
EXCHANGE = Indicator variable set equal to one if the company trades on a major exchange (NYSE, American, NASDAQ), and zero otherwise.
BTM = Book value of equity divided by the market value of equity.
ACQUISITION = Indicator variable set equal to one if the company reports an acquisition, and zero otherwise.
AFILER = Indicator variable set equal to one if the company is an accelerated filer, and zero otherwise.
AFILER_LARGE = Indicator variable set equal to one if the company is a large accelerated filer, and zero otherwise.
C. Cassell et al.
Table 2 Pearson and Spearman correlations

1 2 3 4 5 6 7 8 9 10 11

MISSTATE 1 0.064 0.055 −0.098 −0.067 0.011 0.045 −0.012 0.095 −0.003 0.021
0.000 0.002 0.000 0.000 0.518 0.011 0.503 0.000 0.846 0.228
DELAY 2 0.057 0.260 0.023 0.205 0.084 −0.247 0.182 −0.116 0.210 −0.327
0.001 0.000 0.198 0.000 0.000 0.000 0.000 0.000 0.000 0.000
LATE_FILE 3 0.055 0.300 −0.011 0.021 0.082 −0.015 0.205 −0.016 0.101 −0.134
0.002 0.000 0.551 0.244 0.000 0.409 0.000 0.378 0.000 0.000
BANK_SPEC 4 −0.098 −0.024 −0.017 0.072 0.089 −0.283 −0.035 −0.653 −0.045 −0.113
0.000 0.180 0.347 0.000 0.000 0.000 0.047 0.000 0.011 0.000
CRISIS 5 −0.067 0.192 0.021 0.069 −0.022 0.061 0.124 −0.091 0.268 −0.452
0.000 0.000 0.244 0.000 0.209 0.001 0.000 0.000 0.000 0.000
LEVERAGE 6 −0.036 0.013 0.043 0.184 −0.013 0.110 0.149 −0.037 0.029 −0.181
0.040 0.465 0.016 0.000 0.460 0.000 0.000 0.039 0.105 0.000
CLIENTSIZE 7 0.045 −0.174 −0.041 −0.228 0.048 0.235 −0.084 0.536 −0.120 0.167
A hidden risk of auditor industry specialization: evidence from the...

0.011 0.000 0.020 0.000 0.006 0.000 0.000 0.000 0.000 0.000
GCO 8 −0.012 0.233 0.205 −0.034 0.124 0.124 −0.145 −0.053 0.271 −0.215
0.503 0.000 0.000 0.056 0.000 0.000 0.000 0.003 0.000 0.000
BIG4 9 0.095 −0.051 −0.016 −0.667 −0.091 −0.059 0.513 −0.053 −0.079 0.231
0.000 0.004 0.378 0.000 0.000 0.001 0.000 0.003 0.000 0.000
LOSS 10 −0.003 0.191 0.101 −0.043 0.268 −0.028 −0.179 0.271 −0.079 −0.488
0.846 0.000 0.000 0.014 0.000 0.119 0.000 0.000 0.000 0.000
ROA 11 −0.002 −0.118 −0.109 0.055 −0.073 −0.222 0.171 −0.282 0.035 −0.308
0.912 0.000 0.000 0.002 0.000 0.000 0.000 0.000 0.045 0.000
TENURE 12 0.013 −0.113 −0.022 −0.195 0.003 0.031 0.368 −0.045 0.356 −0.079 0.025
0.479 0.000 0.216 0.000 0.847 0.075 0.000 0.010 0.000 0.000 0.159
909
910

Table 2 (continued)

1 2 3 4 5 6 7 8 9 10 11

INFLUENCE 13 −0.006 −0.012 0.025 0.439 0.048 0.090 0.060 0.018 −0.245 0.034 0.007
0.739 0.498 0.151 0.000 0.007 0.000 0.001 0.299 0.000 0.056 0.684
INTANGIBLES 14 0.043 0.035 0.018 −0.186 −0.004 −0.247 −0.029 0.089 0.150 0.104 −0.114
0.016 0.049 0.305 0.000 0.805 0.000 0.099 0.000 0.000 0.000 0.000
FOR_OPS 15 0.003 −0.017 −0.026 −0.034 −0.026 0.011 0.091 −0.002 0.041 −0.262 0.037
0.877 0.325 0.138 0.054 0.146 0.553 0.000 0.893 0.020 0.000 0.034
L3_ASSETS 16 0.042 −0.111 −0.026 −0.062 −0.409 0.043 0.083 −0.073 0.093 −0.301 0.010
0.019 0.000 0.141 0.001 0.000 0.015 0.000 0.000 0.000 0.000 0.574
BUSY 17 −0.127 0.016 0.029 0.021 0.025 0.105 0.211 −0.011 0.083 −0.026 0.009
0.000 0.359 0.102 0.241 0.156 0.000 0.000 0.523 0.000 0.148 0.604
EXCHANGE 18 −0.013 −0.133 −0.090 −0.110 0.046 −0.047 0.292 −0.187 0.213 −0.151 0.117
0.470 0.000 0.000 0.000 0.009 0.008 0.000 0.000 0.000 0.000 0.000
BTM 19 −0.020 0.140 0.035 0.136 0.444 −0.072 0.008 −0.067 −0.153 0.190 0.021
0.271 0.000 0.051 0.000 0.000 0.000 0.647 0.000 0.000 0.000 0.226
ACQUISITION 20 −0.014 −0.013 −0.016 −0.036 −0.038 0.068 0.156 −0.067 0.067 −0.297 0.063
0.439 0.460 0.365 0.043 0.031 0.000 0.000 0.000 0.000 0.000 0.000
AFILER 21 0.070 0.003 0.015 0.028 −0.015 0.041 −0.030 −0.053 −0.034 −0.033 0.011
0.000 0.882 0.386 0.114 0.384 0.021 0.092 0.003 0.055 0.062 0.528
AFILER_LARGE 22 −0.032 −0.179 −0.044 −0.301 0.019 −0.070 0.615 −0.069 0.466 −0.097 0.076
0.075 0.000 0.014 0.000 0.275 0.000 0.000 0.000 0.000 0.000 0.000
C. Cassell et al.
Table 2 (continued)

Table 2 (continued) 12 13 14 15 16 17 18 19 20 21 22

12 13 14 15 16 17 18 19 20 21 22

MISSTATE 0.013 −0.049 0.037 0.003 0.042 −0.127 −0.013 −0.032 −0.014 0.070 −0.032
0.479 0.006 0.037 0.877 0.019 0.000 0.470 0.073 0.439 0.000 0.075
DELAY −0.168 −0.021 −0.112 −0.046 −0.134 −0.008 −0.179 0.270 −0.022 0.023 −0.298
0.000 0.231 0.000 0.010 0.000 0.655 0.000 0.000 0.213 0.199 0.000
LATE_FILE −0.022 0.039 −0.003 −0.026 −0.026 0.029 −0.090 0.036 −0.016 0.015 −0.044
0.216 0.028 0.878 0.138 0.141 0.102 0.000 0.044 0.365 0.386 0.014
BANK_SPEC −0.184 0.606 −0.189 −0.031 −0.059 0.012 −0.106 0.194 −0.031 0.022 −0.282
0.000 0.000 0.000 0.079 0.001 0.513 0.000 0.000 0.077 0.213 0.000
CRISIS 0.003 0.071 −0.029 −0.026 −0.409 0.025 0.046 0.535 −0.038 −0.015 0.019
0.847 0.000 0.099 0.146 0.000 0.156 0.009 0.000 0.031 0.384 0.275
LEVERAGE 0.014 0.059 −0.293 −0.024 0.029 0.090 −0.095 −0.081 −0.005 0.064 −0.132
0.441 0.001 0.000 0.176 0.101 0.000 0.000 0.000 0.768 0.000 0.000
CLIENTSIZE 0.342 −0.001 0.415 0.052 0.082 0.222 0.298 −0.095 0.129 0.063 0.619
A hidden risk of auditor industry specialization: evidence from the...

0.000 0.962 0.000 0.003 0.000 0.000 0.000 0.000 0.000 0.000 0.000
GCO −0.045 0.027 −0.047 −0.002 −0.073 −0.011 −0.187 0.019 −0.067 −0.053 −0.069
0.010 0.133 0.008 0.893 0.000 0.523 0.000 0.272 0.000 0.003 0.000
BIG4 0.356 −0.476 0.292 0.041 0.093 0.083 0.213 −0.247 0.067 −0.034 0.466
0.000 0.000 0.000 0.020 0.000 0.000 0.000 0.000 0.000 0.055 0.000
LOSS −0.079 0.037 0.005 −0.262 −0.301 −0.026 −0.151 0.219 −0.297 −0.033 −0.097
0.000 0.035 0.779 0.000 0.000 0.148 0.000 0.000 0.000 0.062 0.000
ROA 0.138 −0.100 0.109 0.059 0.172 0.038 0.170 −0.628 0.064 0.056 0.233
0.000 0.000 0.000 0.001 0.000 0.031 0.000 0.000 0.000 0.002 0.000
TENURE −0.140 0.169 0.032 0.044 0.058 0.124 −0.092 0.049 −0.074 0.351
0.000 0.000 0.075 0.014 0.001 0.000 0.000 0.005 0.000 0.000
911
912

Table 2 (continued)

12 13 14 15 16 17 18 19 20 21 22

INFLUENCE −0.071 0.055 −0.016 −0.033 0.071 −0.041 0.118 0.001 0.002 −0.046
0.000 0.002 0.353 0.062 0.000 0.021 0.000 0.936 0.896 0.010
INTANGIBLES 0.079 −0.027 0.010 0.060 0.115 0.177 −0.021 0.137 −0.029 0.368
0.000 0.132 0.578 0.001 0.000 0.000 0.247 0.000 0.096 0.000
FOR_OPS 0.032 −0.039 0.037 0.267 −0.055 0.027 −0.039 0.757 −0.057 0.105
0.075 0.026 0.036 0.000 0.002 0.133 0.030 0.000 0.001 0.000
L3_ASSETS 0.044 −0.025 −0.027 0.267 −0.032 0.023 −0.228 0.260 0.025 0.059
0.014 0.164 0.122 0.000 0.068 0.195 0.000 0.000 0.165 0.001
BUSY 0.058 0.040 −0.030 −0.055 −0.032 0.080 −0.024 0.042 0.046 0.108
0.001 0.025 0.086 0.002 0.068 0.000 0.177 0.017 0.010 0.000
EXCHANGE 0.124 0.001 0.023 0.027 0.023 0.080 −0.083 0.069 0.170 0.291
0.000 0.946 0.202 0.133 0.195 0.000 0.000 0.000 0.000 0.000
BTM −0.044 0.055 −0.165 −0.038 −0.202 0.045 −0.033 −0.030 −0.064 −0.204
0.013 0.002 0.000 0.030 0.000 0.012 0.063 0.093 0.000 0.000
ACQUISITION 0.049 −0.043 0.062 0.757 0.260 0.042 0.069 −0.021 −0.032 0.136
0.005 0.014 0.001 0.000 0.000 0.017 0.000 0.229 0.069 0.000
AFILER −0.074 0.006 −0.044 −0.057 0.025 0.046 0.170 −0.018 −0.032 −0.446
0.000 0.720 0.013 0.001 0.165 0.010 0.000 0.323 0.069 0.000
AFILER_LARGE 0.351 0.021 0.156 0.105 0.059 0.108 0.291 −0.128 0.136 −0.446
0.000 0.241 0.000 0.000 0.001 0.000 0.000 0.000 0.000 0.000

Pearson correlations are in the bottom left. Spearman correlations are in the top right. P-values appear below coefficients. See Table 1 for variable definitions
C. Cassell et al.
A hidden risk of auditor industry specialization: evidence from the... 913

Table 3 The effect of banking auditor industry specialization on bank client audit outcomes during the
financial crisis and pre-crisis periods

(1) (2) (3)

MISSTATE DELAY LATE_FILE

Coeff. t-value Coeff. t-value Coeff. t-value

BANK_SPEC −0.026 * (−1.71) −7.687 *** (−4.00) −0.054 * (−1.88)


CRISIS −0.081 *** (−5.15) 5.652 *** (3.79) −0.101 *** (−4.57)
BANK_SPEC x CRISIS 0.060 *** (4.00) 13.507 *** (8.49) 0.094 *** (3.27)
LEVERAGE −0.065 ** (−1.97) 8.930 ** (2.08) 0.049 (0.97)
CLIENTSIZE 0.013 *** (2.87) −1.865 *** (−5.08) −0.004 (−0.74)
GCO −0.003 (−0.11) 22.184 *** (4.47) 0.359 *** (5.08)
BIG4 0.026 * (1.77) 4.883 *** (3.14) 0.009 (0.45)
LOSS 0.005 (0.41) 2.623 ** (2.01) 0.017 (0.81)
ROA −0.044 (−1.62) 1.321 (0.23) −0.087 (−0.84)
TENURE −0.003 (−0.24) −2.055 ** (−2.16) −0.005 (−0.38)
INFLUENCE 0.025 (0.93) 5.567 ** (2.44) 0.054 (1.56)
INTANGIBLES 0.093 (1.33) 6.177 (1.02) 0.003 (0.03)
FOR_OPS −0.002 (−0.21) −0.312 (−0.30) −0.026 * (−1.73)
L3_ASSETS 0.008 (1.11) −0.277 (−0.26) −0.013 (−0.58)
BUSY −0.072 *** (−3.20) 2.433 (1.46) 0.028 * (1.86)
EXCHANGE −0.014 (−1.27) −2.293 ** (−2.08) −0.039 ** (−2.56)
BTM 0.005 (1.31) 1.336 ** (2.41) 0.016 * (1.68)
ACQUISITION −0.002 (−0.21) 1.688 (1.61) 0.026 * (1.70)
AFILER 0.012 (1.14) −1.916 * (−1.86) 0.026 * (1.66)
AFILER_LARGE −0.039 ** (−2.52) −5.437 *** (−3.43) 0.009 (0.40)
Constant 0.049 (1.48) 65.973 *** (14.56) 0.007 (0.14)
Year FE Yes Yes Yes
MSA FE Yes Yes Yes
Joint Tests:
BANK_SPEC + BANK_SPEC x 0.034 ** (2.08) 5.82 *** (2.72) 0.04 (1.10)
CRISIS = 0
Observations 3199 3199 3199
Adjusted R2 0.086 0.268 0.104

This table presents results from the estimation of Eq. (1), used to test H1. See Table 1 for variable definitions.
Standard errors are clustered by company. *, **, *** indicate significance at p < 0.10, p < 0.05, and p < 0.01
(two-tailed), respectively

specialization variable (high (low) bank specialization is characterized as bank fees


greater (less) than 50% of total office audit fees). Figure 1 shows that, in the pre-crisis
period, high bank specialist auditors had better outcomes than low specialist auditors;
however, this flips in the crisis period such that high bank specialist auditors had worse
outcomes than low specialist auditors.
914 C. Cassell et al.

a
Pre-Crisis Period

b
Crisis Period

Fig. 1 High and low specialization audit outcomes. This figure depicts predicted marginal means for high
bank specialization audit offices compared to low bank specialization audit offices for the pre-crisis and crisis
periods. For this figure, a high (low) bank specialization auditor is one with bank fees greater (less) than 50%
of total office audit fees

Collectively, the results in Table 3 and Fig. 1 provide strong support for H1 and suggest
that the benefits of banking auditor industry specialization for audit outcomes are
mitigated or reversed during the financial crisis compared to the pre-crisis period. We
infer that audit outcomes were adversely affected during the financial crisis because audit
offices with large portfolios of clients from the banking industry faced both a heightened
demand for resources and a limited supply of available additional resources to draw from.

4.2 Additional analyses

In this section, we report results from two sets of tests designed to support our inference
that the observed deterioration in the benefits of banking auditor industry specialization
is driven, at least in part, by difficulties associated with securing and allocating
resources to clients from the banking industry.

4.2.1 Cross-sectional analysis – Audit offices with high and low ROA banks

Our primary tests treat all banks (adjusted for bank size) as requiring (demanding) a
similar amount of additional resources to deal with the risks that arose during the
A hidden risk of auditor industry specialization: evidence from the... 915

financial crisis. In cross-sectional tests, we relax this assumption by assuming that audit
offices with bank clients that are performing relatively well (poorly) will demand less
(more) additional resources, such that the effects documented in Table 3 will be driven
primarily by audit offices with poorly performing banks.
We categorize observations by audit office, where high (low) ROA bank offices are
those where the audit fee-weighted average of ROA for banks in that office is above
(below) the sample median (offices are classified as high or low bank ROA offices on
an annual basis). We then estimate Eq. (1) separately for the sample of banks in high
and low ROA audit offices. The results are presented in Tables 4 and 5.
Table 4 presents results from cross-sectional tests where the dependent variable is
MISSTATE, and Table 5 presents results from cross-sectional tests where the dependent
variable is DELAY. In both tables, Column (1) presents results for the sample of banks
in high ROA audit offices, and Column (2) presents results for the sample of banks in
low ROA audit offices. In both tables, the coefficient on the interaction between
BANK_SPEC and CRISIS is positive and significant in Columns (1) and (2). While
the coefficients on MISSTATE are directionally consistent with our expectations, the
joint tests and the test between high and low ROA offices are not significant at
traditional levels using two-tailed tests. The joint tests in Table 5 (DELAY) indicate
that specialization was detrimental for clients of bank specialist offices with low ROA
clients, and the incremental effect of auditor industry specialization during the crisis
(Column 3) was significantly larger at offices with low ROA banks than at offices with
high ROA banks.29 Overall, these findings provide some evidence suggesting that the
documented effects were stronger at audit offices where the demand for additional
resources was likely to be the greatest.

4.2.2 Other industry clients

Our primary tests do not consider the implications of changes to office-level resource
allocation plans for OICs. If the inferences that we draw from our primary tests are
correct—that is, the demand for (supply of) additional resources is greatest (lowest) in
offices where the client portfolio is weighted towards banking clients—then the effect
of any reallocation of resources should be most pronounced for OICs that engage
banking specialist auditors. Stated differently, while H1 predicts a crisis-induced
deterioration in audit outcomes for banking clients of bank specialists, we also expect
a crisis-induced deterioration in audit outcomes for OICs of bank specialists.30

29
In untabulated analyses, we perform similar tests where the dependent variable is LATE_FILE. Here, the
coefficient on the interaction between BANK_SPEC and CRISIS is positive and significant for offices with
high ROA banks and positive and insignificant for offices with low ROA banks. The test contrasting the
coefficients on the interaction terms for high and low ROA offices is insignificant.
30
We do not make a prediction about the association between audit outcomes and banking auditor special-
ization in the pre-crisis period (main effects) because we have no reason to expect offices with banking
specialization to be better or worse at conducting nonbanking engagements during normal times. That is,
having banking expertise does not preclude an office from having expertise in other industries. Our predictions
about the differential effects of banking audit industry specialization for OICs during the crisis (i.e., the
interaction effects) are identical to those for banking clients because the impacts flow from the same source—
heightened office-level (banking clients and OICs) resource constraints among offices with a high concentra-
tion of banking clients.
916 C. Cassell et al.

Table 4 Cross-sectional tests: offices with high and low ROA banks - MISSTATE

DV: MISSTATE (1) (2) (3)

Offices with High Offices with Low Test of


ROA Banks ROA Banks Differences

Coeff. t-value Coeff. t-value Coeff. Chi2

BANK_SPEC −0.034 * (−1.67) −0.040 * (−1.67)


CRISIS −0.066 *** (−2.99) −0.101 *** (−3.91)
BANK_SPEC x CRISIS 0.048 ** (2.20) 0.078 *** (3.43) 0.030 (0.86)
LEVERAGE −0.107 * (−1.93) −0.078 * (−1.83)
CLIENTSIZE 0.010 * (1.93) 0.018 ** (2.23)
GCO 0.001 (0.04) −0.004 (−0.13)
BIG4 0.037 * (1.76) 0.003 (0.12)
LOSS −0.006 (−0.61) −0.001 (−0.05)
ROA −0.318 ** (−2.12) −0.002 (−0.05)
TENURE 0.003 (0.23) −0.001 (−0.05)
INFLUENCE 0.060 * (1.82) 0.011 (0.28)
INTANGIBLES 0.018 (0.25) 0.165 (1.51)
FOR_OPS −0.005 (−0.36) 0.008 (0.47)
L3_ASSETS −0.004 (−0.37) 0.021 (1.50)
BUSY −0.066 ** (−2.31) −0.086 *** (−3.15)
EXCHANGE −0.026 (−1.65) −0.004 (−0.28)
BTM −0.002 (−0.40) 0.005 (1.02)
ACQUISITION 0.009 (0.64) −0.023 (−1.43)
AFILER 0.021 (1.54) 0.009 (0.53)
AFILER_LARGE −0.030 (−1.53) −0.037 * (−1.77)
Constant 0.115 ** (2.13) 0.000 (0.01)
Year FE Yes Yes
MSA FE Yes Yes
Joint Tests:
BANK_SPEC + BANK_SPEC x 0.014 (0.599) 0.038 (1.486)
CRISIS = 0
Observations 1553 1646
Adjusted R2 0.159 0.107

See Table 1 for variable definitions. Standard errors are clustered by company. *, **, *** indicate significance
at p < 0.10, p < 0.05, and p < 0.01 (two-tailed), respectively

The sample is comprised of company-year observations from 2004 through 2009


with necessary data in Compustat and Audit Analytics to construct the variables in Eq.
(1). We exclude banks (i.e., the observations in our primary sample), utility companies,
and companies from other highly regulated industries because they face different
regulatory and reporting requirements. As in our primary sample, we restrict the OIC
sample to domestic companies with a domestic auditor, and we exclude auditors with
A hidden risk of auditor industry specialization: evidence from the... 917

Table 5 Cross-sectional tests: offices with high and low ROA banks - DELAY

DV: MISSTATE (1) (2) (3)

Offices with High ROA Offices with Low ROA Test of Differences
Banks Banks

Coeff. t-value Coeff. t-value Coeff. Chi2

BANK_SPEC −8.093 *** (−2.73) −7.169 *** (−2.76)


CRISIS 1.303 (0.74) 9.486 *** (3.74)
BANK_SPEC x CRISIS 11.951 *** (5.13) 15.725 *** (6.44) 3.774 *** (28.10)
LEVERAGE 7.464 (0.67) 7.165 (1.47)
CLIENTSIZE −1.457 *** (−3.20) −2.334 *** (−5.12)
GCO 62.949 ** (2.23) 19.945 *** (3.85)
BIG4 2.379 (1.29) 7.137 *** (3.02)
LOSS 0.257 (0.10) 2.441 (1.51)
ROA −26.722 (−0.78) 2.769 (0.46)
TENURE −2.370 ** (−2.35) −0.757 (−0.51)
INFLUENCE 3.350 (1.12) 8.244 ** (2.34)
INTANGIBLES 10.873 (1.55) 5.128 (0.54)
FOR_OPS 0.231 (0.14) −0.505 (−0.40)
L3_ASSETS −1.847 (−1.21) 1.150 (0.68)
BUSY 0.467 (0.20) 3.585 * (1.65)
EXCHANGE −3.549 ** (−2.49) −1.840 (−1.32)
BTM 2.787 *** (2.79) 0.205 (0.31)
ACQUISITION 0.509 (0.29) 2.141 (1.61)
AFILER −2.032 (−1.27) −1.257 (−1.01)
AFILER_LARGE −5.424 ** (−2.53) −3.877 * (−1.78)
Constant 73.660 *** (6.85) 66.577 *** (11.28)
Year FE Yes Yes
MSA FE Yes Yes
Joint Tests:
BANK_SPEC + 3.858 (1.563) 8.556 *** (2.429)
BANK_SPEC x CRISIS = 0
Observations 1553 1646
Adjusted R2 0.264 0.312

See Table 1 for variable definitions. Standard errors are clustered by company. *, **, *** indicate significance
at p < 0.10, p < 0.05, and p < 0.01 (two-tailed), respectively

fewer than ten public clients. The final sample consists of 18,738 OIC company-year
observations.31

31
We include industry-crisis fixed effects in our OIC tests because the impact of the financial crisis likely
varied across industries. We exclude industry-crisis fixed effects from our main tests because the sample is
limited to banks.
918 C. Cassell et al.

Table 6 Other industry clients – MISSTATE and ACCRUALS

(1) (2)

MISSTATE ACCRUALS

Coeff. t-value Coeff. t-value

BANK_SPEC −0.025 (−1.12) −0.007 (−1.08)


CRISIS −0.049 * (−1.73) −0.040 * (−1.88)
BANK_SPEC x CRISIS 0.057 * (1.87) 0.014 (1.30)
LEVERAGE 0.006 * (1.68) 0.001 (0.44)
CLIENTSIZE 0.006 *** (2.94) −0.007 *** (−10.81)
GCO −0.011 (−1.29) 0.020 *** (3.83)
BIG4 −0.009 (−1.25) −0.004 (−1.54)
LOSS 0.006 (1.05) 0.009 *** (4.71)
ROA −0.006 * (−1.77) −0.031 *** (−6.82)
TENURE −0.011 * (−1.81) −0.005 *** (−3.29)
INFLUENCE −0.019 (−1.16) 0.005 (0.90)
INTANGIBLES 0.015 (0.97) −0.021 *** (−4.33)
FOR_OPS 0.008 (1.51) −0.003 (−1.54)
L3_ASSETS −0.008 (−1.62) −0.004 (−1.46)
BUSY −0.013 ** (−2.20) 0.001 (0.63)
EXCHANGE −0.020 ** (−2.23) −0.009 *** (−3.18)
BTM 0.004 ** (2.12) −0.004 *** (−4.32)
ACQUISITION 0.000 (0.09) 0.005 *** (2.94)
AFILER 0.015 * (1.87) 0.000 (0.12)
AFILER_LARGE −0.006 (−0.64) −0.000 (−0.09)
Constant −0.029 (−1.04) 0.151 *** (6.84)
Year FE Yes Yes
Industry-Crisis FE Yes Yes
MSA FE Yes Yes
Observations 18738 17434
R-squared 0.060 0.200

ACCRUALS is the absolute value of discretionary accruals, estimated using the modified Jones model
(including prior year return on assets), following Dechow and Sloan (1995). See Table 1 for all other variable
definitions. Standard errors are clustered by company. *, **, *** indicate significance at p < 0.10, p < 0.05,
and p < 0.01 (two-tailed), respectively

Table 6 presents results where the dependent variable is MISSTATE in Column (1)
and ACCRUALS in Column (2).32 Table 7 presents results where the dependent variable
is DELAY in Column (1) and LATE_FILE in Column (2). Because we do not have a
32
Following Dechow and Sloan (1995), we use the modified Jones model (including prior year return on
assets) to estimate the absolute value of discretionary accruals (ACCRUALS). Tests where ACCRUALS is the
dependent variable use a smaller sample due to data constraints. We do not use ACCRUALS as a dependent
variable in our analyses of bank audits because there are no widely accepted discretionary accrual models for
banks.
A hidden risk of auditor industry specialization: evidence from the... 919

Table 7 Other industry clients – DELAY and LATE_FILE

(1) (2)

DELAY LATE_FILE

Coeff. t-value Coeff. t-value

BANK_SPEC −2.978 (−1.42) −0.028 (−1.32)


CRISIS −4.081 (−0.62) −0.535 *** (−6.13)
BANK_SPEC x CRISIS 10.307 *** (3.03) 0.033 (1.09)
LEVERAGE 1.335 ** (2.57) 0.014 ** (2.49)
CLIENTSIZE −1.480 *** (−6.60) −0.006 ** (−2.45)
GCO 7.027 *** (5.57) 0.159 *** (10.00)
BIG4 2.985 *** (4.16) −0.011 (−1.40)
LOSS 4.690 *** (8.71) 0.032 *** (5.57)
ROA 0.640 (1.31) −0.004 (−0.66)
TENURE −2.950 *** (−5.71) −0.014 *** (−2.60)
INFLUENCE 11.537 *** (5.54) 0.077 *** (3.88)
INTANGIBLES 3.636 *** (2.86) 0.055 *** (3.32)
FOR_OPS 0.480 (0.95) 0.025 *** (4.37)
L3_ASSETS 1.872 *** (3.73) −0.002 (−0.30)
BUSY 0.726 (1.28) 0.057 *** (10.25)
EXCHANGE −4.671 *** (−5.15) −0.067 *** (−6.71)
BTM −0.676 ** (−1.97) −0.005 (−1.36)
ACQUISITION 0.984 ** (2.11) 0.015 *** (2.84)
AFILER −1.230 * (−1.69) 0.010 (1.35)
AFILER_LARGE −7.221 *** (−8.32) −0.005 (−0.59)
Constant 71.613 *** (10.72) 0.116 (1.33)
Year FE Yes Yes
Industry-Crisis FE Yes Yes
MSA FE Yes Yes
Observations 18738 18738
R-squared 0.168 0.141

See Table 1 for variable definitions. Standard errors are clustered by company. *, **, *** indicate significance
at p < 0.10, p < 0.05, and p < 0.01 (two-tailed), respectively

prediction about the main effect of BANK_SPEC (i.e., we do not have a prediction
about the association between audit outcomes for OICs and banking auditor industry
specialization in either period), we focus our discussion on the interaction term. We find
a positive and significant coefficient on the interaction between BANK_SPEC and
CRISIS in two of the four tests. In Table 6, Column (2), the coefficient on the
interaction is not significant at traditional levels using a two-tailed test, but would be
significant using a one-tailed test.33 Collectively, the results in Tables 6 and 7 provide

33
In untabulated analyses, we perform (separate) tests for OICs with estimated positive and negative
discretionary accruals. Here, the signs on the coefficients on the interaction terms between BANK_SPEC
and CRISIS are consistent with expectations. However, the coefficients are not statistically significant.
920 C. Cassell et al.

some evidence suggesting a crisis-induced deterioration in audit outcomes for OICs of


bank specialists.

4.3 Supplemental and robustness tests

4.3.1 Regional analysis

In our main tests, we measure BANK_SPEC at the office level because we expect that
resource reallocation plans are likely to be executed primarily at that level. We
acknowledge, however, that audit offices often share resources with affiliate offices
within their geographic region. This could include the reassignment of auditors, the
reassignment of clients, or a mixture of both.34 Thus, we perform tests after calculating
BANK_SPEC at the (within firm) region level.35 Inferences from the untabulated results
are consistent with those in Table 3.

4.3.2 Small and large auditor analysis

In our main tests, we do not investigate whether our results differ for clients of
large versus small audit firms. Although large audit firms might be better posi-
tioned to mitigate acute resource constraints by shifting resources from other
offices in their own network, smaller audit firms likely have a larger proportion
of (unobservable) private clients that could provide similar resource relief. These
competing predictions limit our ability to draw strong inferences from cross-
sectional tests that exploit differences in audit firm size. Nevertheless, in
untabulated exploratory tests, we estimate Eq. (1) separately for the sample of
banks with Big 4 auditors and the sample of banks with non-Big 4 auditors. In the
tests of cross-equation differences in the coefficients on the interaction between
BANK_SPEC and CRISIS, we find no evidence of differential effects in the two
groups.

4.3.3 Year by year analysis

Our main tests contrast the effects of banking auditor industry specialization
during the financial crisis with the pre-crisis period. This design provides the
cleanest sample in which to test our hypothesis because we can rely on prior
research to predict that auditors will respond to increases in risk symmetrically.
Performing tests that investigate annual effects is problematic because it is much
more difficult to determine: 1) when the crisis period begins (i.e., when the risk
begins to increase), 2) when the post-crisis period begins (i.e., when the risk

34
Consistent with this, our discussions with a partner and a manager who conducted bank audits during the
financial crisis at a Big 4 firm indicated that they frequently lead audits of bank clients from another office
within the same region. They also mentioned that sharing manager and partner resources across offices within
a region was common in the banking practice because of industry expertise.
35
Audit Analytics reports six “auditor region” classifications in the United States: Mid-Atlantic, Midwest,
New England, Southeast, Southwest, and West. In our discussions with an audit manager from a Big 4 firm,
the regional designations from Audit Analytics generally aligned with how the firm divided up the regions.
A hidden risk of auditor industry specialization: evidence from the... 921

begins to subside), and 3) how and when auditors will respond to increases or
decreases in risk (or how their response will translate into observable measures).36
Nevertheless, in untabulated exploratory tests, we examine over-time changes in the
coefficient on BANK_SPEC during the pre- and post-crisis years. In tests where the
dependent variable is MISSTATE, the coefficient on BANK_SPEC increases beginning
in 2007, peaks in 2009, and declines steeply in 2010. In tests where the dependent
variable is DELAY, the coefficient on BANK_SPEC increases beginning in 2006, peaks
in 2011, and declines thereafter. Finally, in tests where the dependent variable is
LATE_FILE, the coefficient on BANK_SPEC increases beginning in 2008, peaks in
2009, and declines sharply in 2010. Collectively, these findings are consistent with a
fairly rapid deterioration in the benefits of banking industry specialization as the crisis
period began, and a gradual recovery in the benefits of banking industry specialization
as the crisis-induced risks subsided.

4.3.4 Market-based measure of auditor industry specialization

In the introduction, we provide several arguments to support our selection of an office-


level measure of banking auditor industry specialization over a measure that relies on a
market-based benchmark. For completeness, we also report the results of tests using a
market-based measure. We measure market-based banking auditor industry specializa-
tion as the sum of audit fees collected from bank clients in the MSA divided by the sum
of audit fees collected from all banking clients by all audit firms in the MSA. The
correlation between the office-level measure of banking auditor industry specialization
and the market-based measure is positive but insignificant and small in economic terms
(0.01). In untabulated analyses, we find a positive coefficient on the main effect of
auditor industry specialization for each of the three dependent variables (the coefficient
is positive and significant when the dependent variable is DELAY) – this suggests that,
during the pre-crisis period, audit outcomes for clients of industry specialists auditors
are either worse or not different than those for clients of nonspecialists. The coefficients
on the interaction terms are insignificant, except when the dependent variable is
DELAY, where it is negative and significant. In sum, the results using a market-based
measure of banking auditor industry specialization indicate either insignificant effects
in both the pre-crisis and crisis periods or effects that are opposite of expectations in
both the pre-crisis and crisis periods.

4.3.5 Robustness tests

We perform a series of additional tests to assess the robustness of our main results.
First, we include (separately) auditor and company fixed effects to alleviate concerns
that our results are attributable to unmeasured auditor- or company-level characteristics.
Second, we replace the raw (unlogged) DELAY dependent variable with the logged
version. Third, we calculate BANK_SPEC using our final sample instead of the full
sample available in Audit Analytics. Fourth, we calculate BANK_SPEC using a three-

36
Our predicted crisis effect will moderate when auditors believe the risks are no longer relevant and reduce
resources committed to those risks or hire and train additional audit personnel sufficient to mitigate sustained
risk.
922 C. Cassell et al.

year window instead of one-year. Fifth, we run our analysis after excluding offices with
fewer than 10 public clients, after excluding offices with more than 75 public clients,
after excluding the five largest banking markets based on the number of banks
(Chicago, Los Angeles, New York, Philadelphia, and Washington D.C.), and after
excluding the five largest banking markets based on total bank audit fees (New York,
Los Angeles, San Francisco, Charlotte, and Washington D.C.). Finally, we adjust the
crisis years to include (separately) 2010 or 2007. Inferences from these untabulated
tests are consistent with those from our main tests (Table 3) in that we observe a
deterioration in the benefits of banking auditor industry specialization during the
financial crisis relative to the pre-crisis period.

5 Conclusion

We investigate the implications of auditor industry specialization on audit outcomes


during times of industry-specific crises. To date, prior literature has provided compel-
ling evidence suggesting that industry specialist auditors provide audits of higher
quality. However, audit researchers and regulators have largely ignored a potential risk
associated with an important aspect of auditor industry specialization—the concentra-
tion of clients in a given industry within the audit office.
We use a measure of office-level industry concentration/specialization (rather than a
market-based measure) and use a difference-in-differences design that focuses on the
effects of banking auditor industry specialization on audit outcomes for bank clients
during and before the financial crisis. We find that, during the period before the
financial crisis, banking auditor industry specialization is associated with higher audit
quality and more timely audits. However, during the financial crisis, the results indicate
that banking auditor industry specialization is associated with lower audit quality and
less timely audits. Results from two sets of additional analyses support our inference
that the observed deterioration in the benefits of auditor industry specialization is driven
by problems associated with securing and allocating resources to clients from the
banking industry. Collectively, our results provide the first empirical evidence suggest-
ing that auditor industry specialization can be detrimental for audit outcomes (audit
quality and audit timeliness) in certain circumstances. Our results should be particularly
concerning because they suggest that the risk associated with auditor industry special-
ization manifests during periods of crisis, when high quality and timely audits are likely
to be particularly important.
Our study is subject to a number of important limitations and caveats. First, our
results are based on observed audit outcomes during a single crisis (the financial crisis
of 2008 and 2009) that had the greatest impact on a single industry (banking). Thus, our
results may not generalize to crises of different magnitudes or crises that primarily
influence other industries. Second, our analyses rely on a measure of auditor industry
specialization that differs from the measure used in most prior work that investigates
the effects of auditor industry specialization. Although we argue that an office-level
measure of auditor industry specialization is warranted in our setting, we acknowledge
that it complicates comparisons between our study and prior work. Third, we assume
that resource allocation plans are centered around availability at the office (in our
primary tests) or region (in supplemental tests) level. We acknowledge that within-firm
A hidden risk of auditor industry specialization: evidence from the... 923

resource sharing could occur outside of those bounds. Fourth, although our difference-
in-differences design and robustness tests (including company fixed effects) should
alleviate concerns about omitted variables, we acknowledge that the financial crisis
resulted in fundamental changes throughout U.S. markets, such that we cannot fully
mitigate concerns related to unobserved covariates. Finally, we acknowledge that our
results do not suggest that auditor industry specialization is detrimental, on average.
Rather, our results suggest that auditor industry specialization can be detrimental in
certain circumstances and that audit firms and audit regulators should consider whether
the audit market, audit firms, or audit offices have become too specialized to handle the
resource allocation problems that crisis situations present.

Acknowledgements We thank the editor (Patricia Dechow), two anonymous reviewers, Christina Adams,
Kris Allee, Ken Bills, Brian Bratten, James Chyz, Jeff Cohen, Jamie Diaz, Andrew Doucet, Fellipe Gomes
Raymundo, Robert Holthausen, Joshua Hunt, Andrew Imdieke, Jadallah Jadallah, Brad Lawson, Bin Li, Linda
Myers, Tom Omer, Evelyn Patterson, Gary Peters, Kathleen Powers, Padmakumar Sivadasan, Jonathan
Shipman, Justin Short, Minlei Ye, seminar participants at the University of Tennessee, seminar participants
at the University of Arkansas, and participants at the 2018 Review of Accounting Studies Conference, the
2017 AAA Southwest Regional Conference, and the 2017 AAA Annual Meeting for helpful comments and
suggestions. We also thank managers and partners of audit firms for several helpful comments. Cory Cassell
gratefully acknowledges financial support from the Ralph L. McQueen Chair in Accounting at the University
of Arkansas. Emily Hunt and Stephen Rowe gratefully acknowledge funding from the Bank of America
summer research grant.

References

Audousset-Coulier, S., Jeny, A., & Jiang, L. (2016). Auditing. A Journal of Practice and Theory, 35(1), 139–
161.
Aaker, D. (1980). Marketing research: Private and Public sector decisions. New York: Wiley.
Andrews, K. R. (1980). The concept of corporate strategy. Homewood, IL: Richard D. Irwin.
Aobdia, D. (2015). Proprietary information spillovers and supplier choice: Evidence from auditors. Review of
Accounting Studies, 20(4), 1504–1539.
Ashton, R. H., Graul, P. R., & Newton, J. D. (1989). Audit delay and the timeliness of corporate reporting.
Contemporary Accounting Research, 5(2), 657–673.
Atiase, R. K., Bamber, L. S., & Tse, S. (1989). Timeliness of financial reporting, the firm size effect, and stock
price reactions to annual earnings announcements. Contemporary Accounting Research, 5, 526–552.
Bajaj, V., & Creswell, J. (2008). A lender failed. Did its auditor? The New York Times.
Balsam, S., Krishnan, J., & Yang, J. S. (2003). Auditor industry specialization and earnings quality. Auditing:
A Journal of Practice & Theory, 22(2), 71–97.
Beardsley, E. L., Goldman, N. C. & Omer, T. C. (2018). Audit office industry diversity and audit quality.
SSRN Working paper.
Becker, C. L., DeFond, M. L., Jiambalvo, J., & Subramanyam, K. R. (1998). The effect of audit quality on
earnings management. Contemporary Accounting Research, 15, 1–24.
Behn, B. K., Searcy, D. L., & Woodroof, J. B. (2006). A within firm analysis of current and expected future
audit lag determinants. Journal of Information Systems, 20(1), 65–86.
Bell, T. B., Doogar, R., & Solomon, I. (2008). Audit labor usage and fees under business risk auditing. Journal
of Accounting Research, 46(4), 729–760.
Berger, P. G., & Ofek, E. (1995). Diversification's effect on firm value. Journal of Financial Economics, 37(1),
39–65.
Berry, C. H. (1975). Corporate growth and diversification. Princeton, NJ: Princeton University Press.
Bhide, A. (1990). Reversing corporate diversification. Journal of Applied Corporate Finance, 3(2), 70–81.
Bills, K. L., Swanquist, Q. T, & R. L. Whited, R. L. (2015). Growing pains: Audit quality and office growth.
Contemporary Accounting Research, 33(1), 288–313.
924 C. Cassell et al.

Bordo, M. D., & Haubrich, J. G. (2009). Credit crises, money and contractions: An historical view. NBER
working paper, 1–46.
Carcello, J. V., & Nagy, A. L. (2004). Client size, auditor specialization and fraudulent financial reporting.
Managerial Auditing Journal, 19(5), 651–668.
Chambers, A. E., & Penman, S. H. (1984). Timeliness of reporting the stock price reaction to earnings
announcements. Journal of Accounting Research, 22(1), 21–47.
Chandler, A. D., Jr. (1962). Strategy and structure. Cambridge, MA: MIT Press.
Christensen, B. E., Glover, S. M., Omer, T. C., & Shelley, M. K. (2016). Understanding audit quality: Insights
from audit professionals and investors. Contemporary Accounting Research, 33(4), 1648–1684.
Coram, P., Ng, J., & Woodliff, D. R. (2004). The effect of risk of misstatement on the propensity to commit
reduced audit quality acts under time budget pressure. Auditing: A Journal of Practice & Theory, 23(2),
159–167.
Cull, R., & Peria, M. (2013). Bank ownership and lending patterns during the 2008–2009 financial crisis:
Evidence from Latin America and Eastern Europe. Journal of Banking & Finance, 37(12), 4861–4878.
Cullen, G., Gasbarro, D., Monroe, G. S., Shailer, G., & Zhang, Y. (2018). Bank audit fees and asset
securitization risks. Auditing: A Journal of Practice and Theory, 37(1), 21–48.
DeAngelo, L. E. (1981). Auditor size and audit quality. Journal of Accounting and Economics, 3(3), 183–199.
Dechow, P. M., & Sloan, R. G. (1995). Detecting earnings management. The Accounting Review, 70(2), 193–
225.
DeFond, M. L. (2010). How should the auditors be audited? Comparing the PCAOB inspections with the
AICPA peer reviews. Journal of Accounting and Economics, 49(1–2), 104–108.
Desai, H., & Jain, P. C. (1999). Firm performance and focus: Long-run stock market performance following
spinoffs. Journal of Financial Economics, 54(1), 75–101.
Doogar, R., Rowe, S., & Sivadasan, P. (2015). Asleep at the wheel again? Bank audits during the financial
crisis. Contemporary Accounting Research, 32(1), 358–391.
Ettredge, M. L., Li, C., & Sun, L. (2006). The impact of SOX section 404 internal control quality assessment
on audit delay in the SOX era. Auditing: A Journal of Practice and Theory, 25(2), 1–23.
Feldman, R., Rosenfeld, B., Lazar, R., Livnat, J., & Segal, B. (2006). Computerized retrieval and classifica-
tion: An application to reasons for late filings with the securities and exchange commission. Intelligent
Data Analysis, 10, 183–195.
Fields, L. P., Fraser, D. R., & Wilkins, M. S. (2004). An investigation of the pricing of audit services for
financial institutions. Journal of Accounting and Public Policy, 23, 53–77.
Francis, J. R., Maydew, E. L., & Sparks, H. C. (1999). The role of big 6 auditors in the credible reporting of
accruals. Auditing: A Journal of Practice and Theory, 18(2), 17–34.
Givoly, D., & Palmon, D. (1982). Timeliness of annual earnings announcements: Some empirical evidence.
The Accounting Review, 57(3), 486–508.
Gluck, F. (1985). A fresh look at strategic management. The Journal of Business Strategy, 6(2), 4–19.
Henderson, C. B., & Kaplan, S. E. (2000). An examination of audit report lag for banks: A panel data
approach. Auditing: A Journal of Practice and Theory, 19(2), 159–174.
Hogan, C. E., & Jeter, D. C. (1999). Industry specialization by auditors. Auditing: A Journal of Practice and
Theory, 18(1), 1–17.
Kinney, W. R., & McDaniel, L. S. (1993). Audit delay for firm correcting quarterly earnings. Auditing: A
Journal of Practice and Theory, 12(2), 135–142.
Knechel, W. R., & Payne, J. L. (2001). Additional evidence on audit report lag. Auditing: A Journal of
Practice and Theory, 20(1), 137–146.
Knechel, W. R., & Sharma, D. S. (2012). Auditor-provided nonaudit services and audit effectiveness and
efficiency: Evidence from the pre- and post-SOX audit report lags. Auditing: A Journal of Practice &
Theory, 31(4), 85–114.
Krishnan, G. V. (2003). Does big 6 auditor industry expertise constrain earnings management? Accounting
Horizons, Supplement: 1–16.
Krishnan, G. V. (2005). The association between big 6 auditor industry expertise and the asymmetric
timeliness of earnings. Journal of Accounting, Auditing & Finance, 20(3), 209–228.
Krishnan, G. V., Sun, L., Wang, Q., & Yang, R. (2013). Client risk management: A pecking order analysis of
auditor response to upward earnings management risk. Auditing: A Journal of Practice & Theory, 32(2),
147–169.
Krishnan, J., & Krishnan, J. (1996). The role of economic trade-offs in the audit opinion decision: An
empirical analysis. Journal of Accounting, Auditing, & Finance, 11(4), 565–586.
Lang, L. H. P., & Stulz, R. M. (1994). Tobin's q, corporate diversification, and firm performance. Journal of
Political Economy, 102, 510–546.
A hidden risk of auditor industry specialization: evidence from the... 925

Lennox, C., & Li, B. (2014). Accounting misstatements following lawsuits against auditors. Journal of
Accounting and Economics, 57, 58–75.
Liu, L., Raghunandan, K., & Rama, D. (2009). Financial restatements and shareholder ratifications of the
auditor. Auditing: A Journal of Practice & Theory, 28(1), 225–240.
Lopez, D. M., & Peters, G. F. (2012). The effect of workload compression on audit quality. Auditing: A
Journal of Practice & Theory, 31(4), 139–165.
Minutti-Meza, M. (2013). Does auditor industry specialization improve audit quality? Journal of Accounting
Research, 51(4), 779–817.
Myers, J. N., Myers, L. A., & Omer, T. C. (2003). Exploring the term of the auditor-client relationship and the
quality of earnings: A case for mandatory auditor rotation? The Accounting Review, 78(3), 779–799.
Norton, E. C., Wang, H., & Ai, C. (2004). Computing interaction effects and standard errors in logit and probit
models. The Stata Journal, 4, 154–167.
Palepu, K. (1985). Diversification strategy, profit performance and the entropy measure. Strategic
Management Journal, 6(3), 239–255.
Palmrose, Z., & Scholz, S. (2004). The circumstances and legal consequences of non-GAAP reporting:
Evidence from restatements. Contemporary Accounting Research, 21(1), 139–180.
Pandya, A. M., & Rao, N. V. (1998). Diversification and firm performance: An empirical evaluation. Journal
of Financial and Strategic Decisions, 11(2), 67–81.
Payne, E. A., & Ramsay, R. J. (2005). Fraud risk assessments and auditors’ professional skepticism.
Managerial Auditing Journal, 20(3), 321–330.
Ponemon, L. A. (1992). Discussant’s response to: Litigation risk broadly considered. In Auditing Symposium
XI: Proceeding of the 1992 Deloitte & Touche/University of Kansas Symposium on auditing problems,
60–72. University of Kansas: School of Business
Pratt, L., & Stice, J. D. (1994). The effects of client characteristics on auditor litigation risk judgments,
required audit evidence, and recommended audit fees. The Accounting Review, 69(4), 639–656.
Public Company Accounting Oversight Board (PCAOB). (2010). Engagement Quality Review. PCAOB
auditing standard no. 7. Washington, DC: PCAOB.
Public Company Accounting Oversight Board (PCAOB). (2015). Concept Release on Audit Quality
Indicators. Release no. 2015–005. Washington, DC: PCAOB.
Reichelt, K. J., & Wang, D. (2010). National and office-specific measures of auditor industry expertise and
effects on audit quality. Journal of Accounting Research, 48(3), 647–686.
Rumelt, R. P. (1982). Diversification strategy and profitability. Strategic Management Journal, 3(4), 359–369.
Shipman, J. E., Swanquist, Q. T., & Whited, R. L. (2017). Propensity score matching in accounting research.
The Accounting Review, 92(1), 213–244.
Simunic, D. A., & Stein, M. T. (1996). The impact of litigation risk on audit pricing: A review of the
economics and the evidence. Auditing: A Journal of Practice and Theory, 15, 119–134.
Solomon, I., & Shields, M. D. (1999). What do industry-specialist auditors know? Journal of Accounting
Research, 37(1), 191–208.
Tan, E. L., & Young, S. M. (2015). An analysis of “little r” restatements. Accounting Horizons, 3(29), 667–
693.
Taylor, M. H. (2010). The effects of industry specialization on auditors’ inherent risk assessments and
confidence judgements. Contemporary Accounting Research, 17(4), 693–712.
Wright, A., & Wright, S. (1997). The effect of industry experience on hypothesis generation and audit
planning decisions. Behavioral Research in Accounting, 9, 273–301.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
926 C. Cassell et al.

Affiliations

Cory Cassell 1 & Emily Hunt 2 & Gans Narayanamoorthy 3 & Stephen P. Rowe 4
1
University of Arkansas, Fayetteville, AR, USA
2
Mississippi State University, Mississippi State, MS, USA
3
Tulane University, New Orleans, LA, USA
4
University of Arkansas, Fayetteville, AR, USA
International Research Journal of Finance and Economics
ISSN 1450-2887 Issue 161 May, 2017
http://www.internationalresearchjournaloffinanceandeconomics.com

The Effect of Discretionary Accruals on Financial Statement


Fraud: The Case of the French Companies

AMARA Ines
Accounting Department
Faculty of Economic and Management Sciences of Sfax, Tunisia
E-mail: amaraines@outlook.com

Abstract

The aim of this paper is to examine the effect of discretionary accruals and
governance mechanisms in the occurrence of financial statement fraud. The sample consists
of 250 annual reports spanning from 2006 to 2010 for listed French companies, which 45
detected fraudulent company by the Financial Market Authority (AMF). Our findings such
that discretionary accruals have a positive effect in corporate fraud, when distinguish
between positive discretionary accruals (aggressive accounting policy) and negative
discretionary accruals (conservative accounting policy); we provide evidence that
aggressive accounting manipulation increases the likelihood of financial statement fraud.
By contrast, conservative accounting policy is negatively associate with corporate fraud.
Additionally, the outside director and ownership concentration are the most significant
variables of governance to explain the corporate fraud.

Keywords: Fraud, Discretionary Accruals, Corporate governance


JEL Classification: G34

1. Introduction
Financial statement fraud is defined as “a deliberate attempt by corporations to deceive or mislead
users of published financial statements, especially investors and creditors, by preparing and
disseminating materially misstated financial statements” (Rezaee, 2005, p. 279). Financial statement
fraud has attracted the attention of the public, press, investors, the financial community, and the
regulators owing to many frauds detected such as Lucent, Xerox, Cendant, Enron, WorldCom,
Adelphia, and Tyco. These multiple financial scandals show a significant difference between the
accounting information disclosed by the managers and the economic reality of companies. According
to the Association of Certified Fraud Examiners (ACFE)1,a corporate fraud worldwide has reached an
amount of $ 3.5 billion in2011. Given this importance, the topic dealing with the determinants of
corporate fraud has attracted a great deal of attention among researchers (e.g. Perols and Lougee, 2011;
Jones et al., 2008). Empirical research dealing with this topic has generally focuses the importance of
governance mechanisms to detect fraudulent companies(Beasley, 1996; Jia et al., 2009; Lisic et al.,
2015).
The corporate governance characteristic such as the board characteristic and the ownership
concentration, have been shown to be effective in reducing the likelihood of accounting fraud and to
ensure the effectiveness and relevance of financial information submit by companies(Dechow et al.,

1The
ACFE is the world's largest anti-fraud organization and premier provider of anti-fraud training and education with more than 75,000
members. The mission of ACFE is to reduce the business fraud and to assist the membership in fraud detection. (http://www.acfe.com).
International Research Journal of Finance and Economics - Issue 161 (2017) 49

2011; Jermias and Gani, 2014).Some researches such as Jones et al. (2008) suggests that discretionary
accruals which can be explain by the legal space that the accounting principles leave for interpretation,
can detect an extreme case of earnings presented as the accounting fraud. According to Dechow et al.
(1996) andHui et al. (2014) find that earnings management is the preliminary step to the violation of
accounting principles generally accepted. Firms can manipulate their financial statements by earning
management using discretionary accruals. Thus, companies with high or less level discretionary
accruals should deal with the consequences of this manipulation, or commit fraud to compensate the
reverse accruals(Beneish et al., 2012; Dechow et al., 2010; Jones et al., 2008).Therefore, the use of
accruals to manipulate results over several years can also push managers to enhance all possible
manipulations(Perols and Lougee, 2011).
It is important to note that access to these irregularities accounting databases are difficult,
consequently the most study dealing with this topic refer usually to U.S. Securities and Exchange
Commission (SEC) enforcement actions to present the irregularity action (Agrawala and Chadha,
2005; Beasley and Carcello, 2000; Crutchley et al., 2007). However, this topic remains under-
researched in European country, especially France where the AMF2 play a critical role to prevent the
fraudulent corporate behavior, improve the functioning of the financial system and, raise the investor
confidence level (Ball and Shivakumar, 2008). In an attempt to fill the gap, we try the choice of France
company, we justify the choice according to recent study in 2014 done by the firm
PricewaterhouseCoopers (PWC) conduct that 55% of French companies reported having experienced
accounting fraud, these results should not be misinterpreted, the increases of fraud it doesn’t
necessarily mean that companies are more affected by this phenomenon but, especially shows that
companies are more likely to detect financial statement fraud.
Given the crucial role played by the AMF to detect the accounting fraud, it becomes important
to understand how different governance mechanisms and management behavior increase or decrease
the likelihood of financial statement fraud in French company. Therefore, the main objective of this
study is to examine the effect of discretionary accruals on financial statement fraud based on a sample
of 250 firms- years observations from the French company indexed CAC All-Tradable over the period
2006–10. In order to make more informative analysis, we examine the effect of the presence and the
sign of discretionary accruals on model fraud (the absolute value, positive discretionary accruals and
negative discretionary accruals).
Our findings show that discretionary accruals have a positive effect on corporate fraud. When
we distinguish between positive discretionary accruals (aggressive accounting policy) and negative
discretionary accruals (conservative accounting policy), we provide evidence that aggressive
accounting manipulation increases the likelihood of financial statement fraud. By contrast,
conservative accounting policies are negatively with corporate fraud. Our findings may help standard-
setters and practitioners identify an important factor that contributes to the reduction of financial
statement fraud and improve the quality of the information quality disclosed to the stakeholders.
The remainder of this paper is organized as follows. Section 2 summarizes prior research
section 3 presents the research hypotheses. Section 4 describes the applied methodology including the
sample, definition of variables and model specification. Section 5 discusses the empirical findings.
Finally, section 6 concludes the paper.

2. Prior Literature
According to Stolowy and Breton (2003) “a legal interpretation can remain true to the spirit of the
standard. It is erroneous, but never fraudulent, this case of earnings management”. Therefore, Fraud
present the same objective as earnings management, but differs from earnings management in that
fraud is outside of generally accepted accounting principles (GAAP), whereas, earnings management is

2
The AMF is an independent public authority that oversees the savings protection and the proper functioning of financial
markets in France.
50 International Research Journal of Finance and Economics - Issue 161 (2017)

within GAAP (Erickson et al. 2006). Given that firms can manipulate financial statements using
accounting practices that are within GAAP or outside of GAAP, we define financial statement
manipulation as occurring when managers commit financial statement fraud or manage earnings (or
both).
Additionally, the sign of the discretionary accruals traces the accounting policy adopted by the
company. In this line,Gietzmann and Ireland (2005) show the different effects of discretionary accruals
signal by breaking down the positive and negative signs. In this context, the research of García.Meca
and Sánchez.Ballesta (2009) through a Meta-analyzes, find that the absolute value accruals and signs
moderate the relation between governance mechanisms and earnings management. Thus, we can
conclude that a company adopts two types of accounting strategies: aggressive accounting strategy and,
conservative accounting strategy.
In the first case, the manager opts for an aggressive accounting to artificially inflate the results,
Therefore, the artificial of the result will push shareholders to demand higher dividends resulting which
leads to a depletion of resources company. For creditors, such a fictitious result can provide assurance
on the operation of the business continuity and its ability to honor its engagements. In addition, such
situation does not reflect the real situation of the company and will induce the users of financial
statements with errors (Kamarudin et al. 2012). So, the earning managements are considered as an
opportunistic behavior of leaders to mislead the use of accounting information in order to appropriate
personal gain or for business needs. Indeed, by engaging in positive discretionary accruals, a rational
manager will be aware of the reversible accruals effect. Face this situation, it committee the fraudulent
adjustments that allow him to make up these financial statements on the profits of shareholders and
investors (Perols and Lougee, 2011).
In the second situation, the manager may choose to conduct for manipulation of down result,
and opted for conservative accounting that tends to underestimate a profit and not to mislead users of
financial statements in error. For example, the level of depreciation increases which be booked in an
accounting period, or overestimating the risk relating to provisions for litigation. In this case, negative
discretionary accruals aim to reduce the level of real income. This conservative strategy will have less
serious effects on the users of financial statements since it allows avoiding overvaluation of its assets,
protecting company assets against exhaustion risk, and reducing the expropriation risk of company’s
resources. When manager opts for negative discretionary accruals aimed at reducing the level of actual
result, this has less serious consequences on resources business and its ability to honor these
commitments.
According to Karjalainen (2011), the company adopts a conservative approach to avoid
overvalued risk of assets against the creditors and the shareholders. This allows to conclude that
opportunistic managers make use of manipulation resulting in discretionary accruals and to hide their
poor performance to better conceal their fraudulent activities. These results were also highlighted by
Perols and Lougee (2011) assert that when companies seek to achieve analysts forecasts, are more
likely to use earnings management for previous years and tend to commit fraud.

3. Hypotheses
For studying the relation between discretionary accruals and financial statement fraud, we test the
following hypotheses:
H1: There is a significant relation between the discretionary accruals and the likelihood of
financial statements fraud.
H1.a. The positive discretionary accruals (aggressive accounting) increase the likelihood of
financial statement fraud.
H1.b The negative discretionary accruals (conservative accounting) decrease the likelihood of
financial statement fraud.
International Research Journal of Finance and Economics - Issue 161 (2017) 51

4. Research Method
4.1 Models
To examine the effect of the overall discretionary accruals on the accounting fraud model, a logit data
analysis is performed for a sample of 250 observations spanning from 2006 to 2010.Our first model
(denoted as Model 1) is in line with several empirical investigations concerning the same issue
including factors related to the motivations and opportunities, leading to performing fraud in financial
statements. It is as follows:
FRAUD it = β0 + β1DAccit + β2SIZE-CAit + β3INDEPit + β4COMPETENCEit +
β5DUALITY it + β6 OWN it + β7 SIZE it + β8ROA it + β9LEVERAGE it +
β10 GROWTHit + si (Model 1)
FRAUD : Binary variable coded 1 if there is fraud in the financial statements, and 0 otherwise,
DAcc, Value of discretionary accruals for firm i in year t as measured by the model of Kothari et al.
(2005), SIZE-CA : Board Size of the firm i in year t, INDEP: Report of the number of outside
directors to the total director’s number, COMPETENCE: Dummy variable equal to 1 if a firm has at
least one director who has a professional qualification in accounting or finance, and 0 otherwise,
DUALITY: Dummy variable equal to 1 if the Chairman of the Board of Directors firms i in year t is
also its CEO, OWN: Level of managerial ownership firm i in year t, SIZE: Size of firm i in year t,
measured by the natural logarithm of total assets, ROA: Ratio of income before extraordinary items to
total assets, LEVERAGE: Debt level of firm i in year t. Ratio of total debt to total assets, GROWTH:
Revenu Growth = (Revenu t –Revenu t−1) /Revenu t−1, 𝗌i : The residual value.
In order to improve our first model tests of the impact of the discretionary accruals measure, we
have applied a refined analysis. So, after testing the effect of different variables in the first model
(Model 1), we eliminate this variable in the second model (Model 2) to review its explanatory power in
relation to other variables in the fraud explanation.
FRAUD it = β0 + β1SIZE-CAit + β2INDEPit + β3COMPETENCEit + β4DUALITY it +
β5OWN it + β6 SIZE it + β7ROA it + β8LEVERAGE it + β9 GROWTHit + si (Model 2)
Furthermore, the sign of discretionary accruals traces the accounting policy of the company.
Therefore, in the fifth and sixth model we separate régressions depending on accruals sign and, try to
study the effect of the accounting policy on the likelihood to commit fraud. To do that, we distinguish
between companies that take a conservative accounting strategy (AD<0) (Model 3) and companies
engaged in aggressive accounting strategy (AD> 0) (Model 4).
FRAUD it = β0 + β1 Positive-DAccit + β2SIZE-CAit + β3INDEPit +
β4COMPETENCEit +β5DUALITYit + β6 OWN it + β7 SIZE it + β8ROA it +
β9LEVERAGE it + β10 GROWTH it + si (Model 3)
FRAUD it = β0 + β1 Negative-DAcc it + β2SIZE-CAit + β3INDEPit + β4COMPETENCEit +
β5DUALITY it + β6 OWN it + β7 SIZE it + β8ROA it + β9LEVERAGE it +
β10 GROWTHit + si (Model 4)
Finally, Warfield et al. (1995) found that absolute discretionary accruals lean to measure the
extent to which managers knowingly pursue techniques to adjust reported numbers. While these
measures ignore the effect of the sign of the accruals and the accounting strategy. Therefore, we will
examine the effect of the absolute value of discretionary accruals on fraud. In fact, the absolute value
of discretionary accruals is a measure of the magnitude of discretionary accruals, irrespective of their
signs (Model 5). In addition, to test the moderating effect of discretionary accruals, we include a
dummy variable, which equals 1 if the company commits accounting fraud and 0 otherwise to test the
effect of discretionary accruals only for fraudulent companies (Model 6).
FRAUD it = β0 + β1 ABS-DAccit + β2SIZE-CAit + β3INDEPit +
β4COMPETENCEit + β5DUALITY it + β6 OWN it + β7 SIZE it + β8ROA it +
β9LEVERAGE it + β10 GROWTHit + si (Model 5)
52 International Research Journal of Finance and Economics - Issue 161 (2017)

FRAUD it = β0 + β1 Dummy-ADit + β2SIZEit + β3INDEPit + β4COMPETENCEit +


β5DUALITY it + β6 OWN it + β7 SIZE it + β8ROA it + β9LEVERAGE it +
β10 GROWTHit + si (Model 6)

4.2 Dependent Variable


Several researchers (e.g Albrecht et al., 2008; Albring et al., 2013; Skousen and Wright, 2006) find that
the binary variable indicating the presence of fraud is the single most statistically significant and can
explain some of total earnings manipulation regularization; This variable is dichotomous and
qualitative. It takes the value 1 if the firm is a victim of fraud in the financial statements; otherwise it is
set to 0.
FRAUD = 1 if the firm has committed fraud in the financial statements.
FRAUD = 0 otherwise

4.3 Test Variables


4.3.1. Discretionary Accruals
Studies show that managers use discretionary accruals as a way to respond to asymmetric information
and agency costs in their private information and strategies for financial information, Choi et al. (2011)
and Hasnan et al. (2012). In this regard, in our research of Jones et al. (2008), we evaluate the ability
of discretionary accruals model to detect the existence of fraudulent events, in fact through the study of
Dechow et al. (1996) it appears that the modified Jones (1995) model is the best, and provides the most
reliable estimate of accruals. Kothari et al. (2005), performs a simulation to evaluate the power of
Jones model and the modified Jones model and finds that with the measures of discretionary accruals
performance improves the conclusions reliability.
1 𝑃𝑃𝐸i𝑡
𝐴𝐶𝐶𝑅i𝑡 +𝛼 ( ) + 𝛼 Δ𝑅𝐸𝑉i𝑡 − Δ𝐴𝑅i𝑡 ( ) + 𝛼 𝑅𝑂𝐴 + s (1)
=𝛼 2( )+𝛼 4 i𝑡 i𝑡
𝑇𝐴i𝑡−1 0 1 𝑇𝐴i𝑡−1 𝑇𝐴i𝑡−1 3 𝑇𝐴i𝑡−1

ACCR it: Accruals for firm i in year t, TAit: Total assets for firm ia t end year t-1, ∆ REVit:
Revenues in year t less revenues in year t-1 for firm I, ∆ ARit: Changes in accounts receivable, PPEit
:gross property, plant, and equipment;Property for firm i at end year t, ROAit: Return On Assets, εit
:Error term for firm j in year t.
The measure of discretionary accruals (DAcc) is the residual of equation (1), It is the difference
between actual total accruals (ACCR) deflated by total assets (TAit) and normal accruals estimated by
the fitted values of equation (1). The measure of actual total accruals (ACCR) is the difference between
net income before extraordinary items and operating cash flows from the statement of cash flows.
𝐴𝐶𝐶𝑅i𝑡 𝑃𝑃𝐸i𝑡
𝐷𝐴𝑐𝑐 = − [𝛼 + 𝛼1 ( 1 ) + 𝛼2 (Δ𝑅𝐸𝑉i𝑡 − Δ𝐴𝑅i𝑡) + 𝛼 ( ) + 𝛼 𝑅𝑂𝐴 ]
0 3 4 i𝑡
𝑇𝐴i𝑡−1 𝑇𝐴i𝑡−1 𝑇𝐴i𝑡−1 𝑇𝐴i𝑡−1
In this context, the quality of financial statements is usually linked to the level of discretionary
accruals. A recent study by García.Meca and Sánchez.Ballesta (2009) show the different effects of
discretionary accruals signals and the governance mechanisms. So, the signal of discretionary accruals
reflects the behavior of the leader towards the expected result. Therefore, in order to improve our
model, we distinguish between companies take a conservative accounting strategy (DAcc < 0) and
companies engaged in aggressive accounting strategy (DAcc > 0).

4.2 Control Variable


Agency theory suggests that there conflicts of interest and information asymmetry between
management and shareholders, these conflicts can explain the corporate fraud (Berle and Means,
1932). Additionally, Empirical studies find that internal control can reduce the probability of frauds
(future) (Jensen, 1993), so the weakness of the governance system is an important factor to determinine
financial statement fraud. In this section, we review the studies that have analyzed the relationship
International Research Journal of Finance and Economics - Issue 161 (2017) 53

between governance mechanisms as the board Characteristic (e.g. board size, outside director, CEO
duality, and competence of administration) and the ownership structure.
The control variables included also the financial business characteristics likely affect financial
statement fraud: firm performance, firms leverage, firm size and growth.

4.4.1 Corporate Governance Variables


Agency theory suggests that board characteristics can improve the corporate reporting policy through
more voluntary disclosure and transparency (Fama and Jensen, 1983). Board size may play an
important role in promoting corporate transparency and in directors’ ability to monitor and control
managers since it increases the pool of expertise and opinions and consequently reduce the probability
of manipulation (Jensen, 1993). In addition, when the board dominated by outside members, it ensures
the proper functioning and increases the relevance and information reliability provided in the financial
statements and consequently reduces the likelihood of financial statements fraud. This is proved by
several researchers (Beasley, 1996; Fich and Shivdasani, 2007; Lennox and Pittman, 2010).
In other hand, the agency theory shows that the separation between the function of Chief
Executive Officer and chairman of the board help the commission to more effectively monitor
managers. Therefore, the CEO duality increases the likelihood of financial statements fraud(Carver,
2014).Additionally, the competence of the administrators is an important factor in ensuring the proper
functioning of the Board and its effectiveness (Carcello, 2009). Therefore, there is an important reason
to examine the relation between financial expertise and information quality.
The competence of administrations in accounting or in finance allows them to have higher
quality monitoring and a better understanding to evaluate the situation of the firms (Xie et al., 2003).
The ownership concentration is the proportion of stock controlled by the majority of owners. High
level of ownership concentration reduces the transparently and relevance of the information (Lim et al.,
2014).Chen et al. (2006) find that ownership structure has an important effect on a firm’s propensity to
commit fraud. Nevertheless, Beasley et al. (2010), find no significant effect between the percentage of
shares held by the majority shareholders and the presence of fraudulent accounting.

4.4.2 Financial Business Characteristics


Firm’s financial business characteristics may include financial performance, debt ratio, growth ratio
and corporate size. Companies are more likely to manipulate accounting figures when performance is
relatively low. For instance, Lisic et al. (2015), Caskey and Hanlon (2013) and Wuerges and Borba
(2010) state that high-profitability firms have greater incentives to signal the quality of their
performance and their ability to manage risks successfully. Furthermore, Jones et al. (2009) consider
that debt ratio may affect accounting manipulations since it describes the financial structure and
measures the long-term risk. Thus, companies are more likely to manipulate earnings to avoid
contravention of the clauses of debt contracts (Defond and Jiambalvo 1991; Velikonja, 2011).
In addition, the third variable of financial business characteristics is the growth. Empirical
evidence asserts that minimizing the fraud risk is associated with the growth stability or benefice
volatility (Albring et al., 2013; Brazel et al., 2009). Perols and Lougee (2011) assert that the financial
stability of the company is related to abnormal fluctuations in the growth rate either upward or
downward. Consequently, higher growth rates may increase the likelihood of financial misstatement.
Finally, large firms are characterized by greater levels of public visibility. Therefore, the empirical
evidence on the effect of firm size and corporate fraud affirms that firm size increase the likelihood of
financial statements fraud (Fich and Shivdasani, 2007; Lennox et al., 2013; Lisic et al., 2015).
Similarly, Watts and Zimmerman (1990) and Cassell et al. (2014) found a significant and positive
relationship between firm size and the level of discretionary accruals and they asserts that firms with a
large size engage more frequently in the earnings management to avoid the political costs.
The independent variables related the governance mechanisms are summarized in table1.
54 International Research Journal of Finance and Economics - Issue 161 (2017)
Table 1: Variables definitions

Abbreviation Sign Variable Definition Data source


Board Characteristics
SIZE-CA + Size of Board Board Size of the firm i in year t, measured The sanctions
by the natural logarithm of the total publications on the
number of directors AMF website
INDEP - Control by Report of the number of outside directors Annual reports from
outside to the total directors’ number. the Euronext
directors

COMPETENCE - The Dummy variable equal to 1 if a firm has at Website of firms


competence of least one director who has a professional
members of the qualification in accounting or finance and
Board 0 otherwise
DUALITY + Dual function Dummy variable equal to 1 if the
Chairman of the Board of Directors firms i
in year t is also its CEO.
OWN + Ownership Level of managerial ownership of the firm
structure i in year t

Financial business characteristics


SIZE + Size of firms The size of firm i in year t, measured by Annual reports from
the natural logarithm of accounting of total the Euronext
assets.
ROA - Performance Ratio of income before extraordinary items
to total assets
LEVERAGE + Leverage The debt level of firm i in year t. Ratio of
total debt to total assets
GROWTH + Growth Revenu Growth = (Revenu t –Revenu
t−1)/Revenu t−1

4.5 Sensitivity Tests


Piot and Janin (2007) use the absolute values of abnormal accruals to measure the overall propensity of
earnings management. According to Warfield et al. (1995), absolute abnormal accruals lean to
determine the level to which managers knowingly pursue techniques to adjust reported figures.
Therefore, we examine the relationship between absolute of abnormal accruals on the fraud model. In
this regard, García.Meca and Sánchez.Ballesta (2009) assert that the sign of the discretionary accruals
model (absolute versus signed) moderates the relation between corporate governance variables and
earnings management. Furthermore, we use a dummy variable which equals 1 if the company
committed accounting fraud and 0 otherwise to test the effect of discretionary accruals only for
fraudulent companies.
In addition, according to Gietzmann and Ireland (2005) we distinguish between firms adopting
aggressive accounting policies from those adopting conservative policies. We do this by adopting a
modified Jones (1991) model to estimate discretionary accruals. This captures the net effect of a
multitude of accounting choices, and is less subjective than constructing a single dichotomous variable
by inspecting and classifying descriptions of accounting policies. Hence, firms with aggressive
accounting policies are defined as those making positive discretionary accruals (Model 4), whereas
those making negative discretionary accruals are defined as conservative (Model 3).
International Research Journal of Finance and Economics - Issue 161 (2017) 55

4.6 Scope of the Research


The findings of this paper have policy implications for practitioners and regulators to reduce the
likelihood of corporate fraud, given the crucial role played by discretionary accruals as an important
indicator to determine accounting fraud. This research is the first in France to focus on how
discretionary accruals and governance mechanisms are evident in fraudulent companies sanctioned by
the financial market authority (AMF).

4.7 Sample and Data Sources


The sample of this research consists of French company’s groups belonging to the CAC All-Tradable
and having adopted international accounting standards from 2005. The period runs for five years
during the period of 2006 to 2010. The information collection about companies that commit fraud in
their financial statements is performed manually from the document databases published by the
newsletters of the AMF “Financial Markets Authority” and they recover based on Euronext and the site
management of legal and administrative information. As part of this research, we use the French CAC
All-Tradable companies to collect data to validate our assumptions and we exclude banks, insurance
companies, and financial institutions, in general, as they are subject to specific rules on accounting. We
have subsequently, extracted accounting information of 45 fraudulent companies over the period 2006-
2010. The selection of control firms is based on the study of Beasley (1996).
The final sample consists of 250 French firm-year observations.

Table 2: Sample selection

Sample obtaining procedure


Number of companies sanctioned by the AMF that have commit
accounting fraud during the period 2006-2010 85
Less financial companies, insurance and investment companies 19
Fraudulent Companies
Less companies do not belong to the CAC ALL Tradable 13
Firms- Years
Less property companies 6
Less companies in which certain information was not available 2
Total Sample 45
More control companies 205 Control companies
Data over 5 years (number of observations) 250

5. Empirical Results
5.1 Descriptive Statistics
Descriptive statistics show that discretionary accruals with amean of (-13.5%) have a maximum of
86% and a minimum of 52%. These results cooraborate with the findings of Jones and al. (2008), in
another context than France. In addition, the board size presents a mean equal to (2.4068) members, it
is relatively high compared to the size considered optimal by Jensen (1993) with a level of seven to
eight members. Moreover, the percentage of independent board, has an average of 49% like found in
the study of Cassell et al.(2014). Additionally, the majority of companies in the sample has a dual
structure of 62% for fraudulent firms, and it’s more important and higher than non-fraudulent firms
38%. These results corroborate with studies of Carver (2014) and Hasnan et al. (2012) to show that,
there is more duality in the role of CEO duality in fraudulent firms, the same result found in both ratios
of quality and competence.The ownership structure presents a mean of 34% and 84% of companies
which are majority owned. These results corraborate those with studies conducted so far on the
ownership structure in France, French institutional ownership is on mean 25.31%.
Concerning the financial characteristic presented by the company ROA variables with amean of
5.13% and the variable leverage with a mean of 18.2%, and the variable growth of 21.27%. The firm
56 International Research Journal of Finance and Economics - Issue 161 (2017)

size in our sample has a mean of 16.06, this result shows a slight increase compared to that found by
Hui et al. (2014).

Table 3: Descriptive statistics

Descriptive statistics for continuous independent variables (N=250)


Variable Mean Sd Variance MAX MIN
DAcc -0.013 0.105 0.011 0.524 -0.867
SIZE-CA 2.406 0.346 0.119 2.890 1.386
INDEP 0.441 0.229 0.052 1 0
OWN 0.364 0.237 0.563 0.844 0.026
ROA 0.513 0.783 0.006 0.571 -0.253
LEV 0.182 0.152 0.023 0.544 0.001
SIZE 16.064 1.508 2.274 20.772 12.487
GROWTH 0.212 1.341 1.799 16.339 -0.998
Descriptive statistics for dummy independent variables (N=250)
Frequency Percentage Percentage cumulative
DUALITY 0 95 38.000 38.000
1 115 62.000 100.00
COMPETENCE 0 40 16.000 16.000
1 210 84.000 100.000
Notes: DAcc : Value of discretionary accruals for firm i in year t as measured by the Kothari et al. (2005) model; SIZE-
CA: Board Size of the firm i in year t, measured by the natural logarithm of the total number of directors; INDEP:
Report of the number of outside directors to the total directors’ number; COMPETENCE: Dummy variable equal
to 1 if a firm has at least one director who has a professional qualification in accounting or finance and 0
otherwise; DUALITY: Dummy variable equal to 1 if the Chairman of the Board of Directors firms i in year t is
also its CEO; OWN: Level of managerial ownership of the firm i in year t; SIZE: The size of firm i in year t,
measured by the natural logarithm of accounting of total assets; ROA : Ratio of income before extraordinary items
to total assets; LEV: The debt level of firm i in year t. Ratio of total debt to total assets; GROWTH: Revenu
Growth = (Revenu t –Revenu t−1)/Revenu t−1.

5.2 Univariate Analysis


Before testing the multivariate analysis, it is necessary to verify the absence of a multicollinearity
problem. Gujarati (1988) noted a strong correlation that allows bias estimates of regression. To verify
the presence or absence of multicollinearity between the explanatory variables, we calculate the
'Variance Inflation Factors "(VIF), the Pearson correlation coefficients and values of tolerance. To
conclude, all VIF have a value less than 10 and the values of tolerance all surplus 0.25 (Myers, 1990).
In addition, Table 4 reports bivariate statistical correlations between all variables. The analysis shows
that fraud is positively and significantly correlated with discretionary accruals (0.445; p= 0.000),
ownership concentration (0.367; p=0.000) and ROA (0.253; p=0.000), and significantly negative
associated with outside board members (-0.543, p=0.000). Such result confirms that high independence
members provide less control, thus, more probability of accounting fraud.
Table 4: Pearson Correlation Matrix
Fraud DAcc SIZE-CA INDEP DUALITY COMPE OWN ROA LEV SIZE GROWTH
Fraud 1.000
DAcc 0.445* 1.000
0.000
SIZE-CA -0.106 0.007 1.000
0.093 0.904
INDEP -0.543* 0.239* 0.163* 1.000
0.000 0.000 0.009
DUALITY 0.023 0.030 0.198* -0.126* 1.000
0.710 0.632 0.001 0.045
COMPET 0.005 -0.091 -0.022 0.111* 0.107 1.000
International Research Journal of Finance and Economics - Issue 161 (2017) 57
Fraud DAcc SIZE-CA INDEP DUALITY COMPE OWN ROA LEV SIZE GROWTH
0.928 0.150 0.721 0.079 0.088
OWN 0.367* 0.130* 0.099* -0.234* -0.113* -0.086 1.000
0.000 0.038 0.116 0.000 0.073 0.171
ROA 0.253* 0.417* -0.000 -0.204* 0.004 -0.093 0.248* 1.000
0.000 0.000 0.996 0.001 0.949 0.141 0.000
LEV 0.081 0.115* -0.030 -0.129* 0.194* 0.000 -0.033 0.035 1.000
0.201 0.687 0.627 0.040 0.002 0.989 0.600 0.579
SIZE 0.013 0.053 0.197* 0.086 -0.089 0.025 0.006 -0.17* -0.133* 1.000
0.833 0.399 0.001 0.175 0.160 0.688 0.916 0.006 0.034
GROWTH -0.067 0.053 -0.070 -0.007 -0.025 0.029 -0.127* 0.016 0.100 0.013 1.000
0.286 0.399 0.268 0.909 0.684 0.646 0.044 0.796 0.114 0.836
Notes: FRAUD: Binary variable coded 1 if there is fraud in the financial statements, and 0 otherwise; DAcc: Value of
discretionary accruals for firm i in year t as measured by the kothari et al. (2005) model; SIZE-CA: Board Size of
the firm i in year t, measured by the natural logarithm of the total number of directors; INDEP: Report of the
number of outside directors to the total directors’ number; COMPETENCE: Dummy variable equal to 1 if a firm
has at least one director who has a professional qualification in accounting or finance and 0 otherwise; DUALITY:
Dummy variable equal to 1 if the Chairman of the Board of Directors firms i in year t is also its CEO; OWN:
Level of managerial ownership of the firm i in year t; SIZE: The size of firm i in year t, measured by the natural
logarithm of accounting of total assets; ROA : Ratio of income before extraordinary items to total assets; LEV:
The debt level of firm i in year t. Ratio of total debt to total assets; GROWTH: Revenu Growth = (Revenu t –
Revenu t−1)/Revenu t−1. Significant at 10%.

5.3 Multivariate Analysis


Table 5 displays the multivariate regression results; we test the effect of the discretionary accruals
presence on the model of determinants of fraudulent information. We find in our results, that the best
performing model is the one who integrates the discretionary accruals variable; this is explained by a
hight R- squared. In fact, the first regression (Model 1) means to test the effect of discretionary
accruals and the corporate governance on the model of accounting fraud. The explanatory power of
that model is quite highly adjusted with R- squard equal to (74.41). In addition, the results find that
accruals variable is the main determinant of fraudulent financial statements, this variable has a positive
and significant effect with a coefficient of (47. 988) and statistical t-Student (3.53). As a result, most of
companies in our sample are engaged in technical results of aggressiveness in their financial statement.
In the second regression (Model 2) when eliminating the variable of discretionary accruals, the
R-squared found value (0.6162) is lower compared to the model 1. Our results are robust and show the
effect of the main component of financial statement fraud. This result corroborates with studies of
Perols and Lougee (2011). In this line, Dechow et al. (1996) show the significant relation between
discretionary accruals and the likelihood of financial statement fraud. In general, our results show that
most of governance mechanisms variables such (board size, competence, cumulative) are insignificant.
This result is not shocking, through a recent study of Brown et al. (2011) and Beasley et al. (2010),
explain that it is difficult to identify the firms most likely to engage in fraud by governance mechanism
since most companies use the same governance rules.
However, the variables of outside directors and ownership concentration are significant in most
models. This implies that, outside director reduces the likelihood of fraudulent financial statements.
Similarly, for the variable ownership concentration increases the probability of making fraudulent
financial statements. Furthermore, variables related to financial structure of firms such as the firms’
size, debt and performance are insignificant in all models, whereas, the variable “Growth” has a
negative sign and it significantly. We can affirm that more companies have highest growth
opportunities, more increase in transparency towards external investors are involved. Consequently, the
likelihood of fraud in the financial statements is reduced. Daske et al. (2008), show that companies
with strong growth opportunities have more investment’s chances and require more external financing
resources. Therefore, it needs funding to create incentives to improve the quality and relevance of
'published accounting information.
58 International Research Journal of Finance and Economics - Issue 161 (2017)

5.4 Additional and Sensitivity Tests


To know the moderating effect of discretionary accruals, we test the effect of the absolute value of
discretionary accruals on the fraud model (Model 5); In fact, the absolute value is a measure of the
magnitude of discretionary accruals, irrespective of their signs. According to tests conducted
discretionary accruals in the financial statements are positively associated to act in an illegal behavior,
this is proven by a significant value of adjusted R² (0.6394). Indeed, the coefficient of discretionary
accrual has a positive sign (7.323) and a Student's t-value (2.54). Excluding signs of discretionary
accruals, we observe that the significant variable in this model is the “presence of outside directors”
with a coefficient of (-18.311) and a Student's t-value (-5.30), the same result, is found for the
concentration of ownership among the majority shareholders, this coefficient is (6.381) with a t-
Student estimated (3.22). This implies that, the abstraction of discretionary accruals signs improves the
probability of detecting fraudulent information.
In the sixth regression (Model 6), we test the effect of discretionary accruals only for
fraudulent companies. Thus, we introduce a binary variable equal to 1 if the company is committed the
fraud and 0 otherwise. According to the results found, the discretionary accrual is significantly
associated with illegal act. This proves a very significant value of R- squared (0.7388), besides the
model does not raise the problem of multicollinearity when the maximum VIF is given value (1.37).
Indeed, the coefficient of discretionary accruals is a positive sign (62.963). Therefore, most fraudulent
companies are engaged on the aggressiveness accounting techniques. Also, concerning the variable
"Growth" measured by the change in turnover, this variable is always significant in our models enacted
with a coefficient (-3.587) and Student's t (-2.22). This corroborates with the results found by Cassell et
al. (2014) which prove that reducing the risk of fraud is less likely when companies have great
potential growth or have more volatile earnings.
Previous studies test only the effect of an aggressive accounting policy and accounting fraud
(Kamarudin et al. 2012), or the effect of conservative policy (Wakil and Alam, 2012). Our research is a
pioneer in this field because it encompasses all the signs of accruals given the absolute value positive
accruals and negative accruals. To prove the test, we will study the effect of conservative accounting
policy, measured by negative discretionary accruals on the likelihood of accounting fraud. Through the
tests, we found that the R-squared is a weak value (30.70%) with a coefficient of discretionary accrual
(-10.11) and a Student’s t of (-2.83). This implies that the variable of negative discretionary accruals
has a significant effect to commit fraud in the financial statements. Similarly, the variable “Board
Size”, which has a negative and significant effect on the likelihood to make accounting fraud. Indeed,
the coefficient is (-2.025) with a Student’s t (-2.08).
We can conclude that the negative discretionary accruals model, which tends to increase
expenses and reduce product has an insignificant effect on the likelihood of fraud since the leader
respect the precautionary principle. In this regard, this policy, which tends to reduce the accounting
results, can lead to a tax adjustment. The latter is qualified as a tax evasion which has less effect on the
reaction of investors.
For the fourth model, we tested only the effect of positive discretionary accruals on the
likelihood of financial statements fraud. The R-squared value found is (67.77%) and the coefficient of
the discretionary accruals is a positive value (53.254) with an estimated Student t- (2.30). In addition,
the independence of board is significant with a coefficient (14,198) and t-Student (-2.85), consequence,
when there are more members of an independent board, the quality of information will be more
relevant and transparent. Similarly, for the variable of ownership concentration among the majority
shareholder, with a coefficient of (6.053) and a t-Student value (2.21), the high percentage of control
block owned by shareholders helps to increase the likelihood of fraudulent financial statements. We
can conclude that the positive discretionary accruals, which tend to increase revenues and reduce costs,
have a significant effect on the likelihood to commit fraud as the leader artificially manipulate profits
rising.
The table below discussion highlights the following key findings:
International Research Journal of Finance and Economics - Issue 161 (2017) 59
Table 5: Logistic Regression

Dependant Variable: Fraud


Sensitivity tests
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Independent
variables Overall DAcc Without DAcc DAcc < 0 DAcc > 0 ABS_ DAcc DUM_ DAcc
Coeff t-statistic Coeff t-statistic Coeff t-statistic Coeff t-statistic Coeff t-statistic Coeff t-statistic
Constant -4.546 -1.11 -4.844 -1.43 -3.455 -0.92 -6.430 -1.11 -4.154 -1.23 -6.674 -1.63
Overall DAcc 47.988*** 3.53
Without DAcc
ABS_ DAcc 7.323*** 2.54
DUM_ DAcc 62.963*** 3.59
DAcc < 0 -10.113*** -2.83
DAcc > 0 53.254*** 2.30
SIZE _CA -0.286 -0.33 0.280 0.36 -2.025** -2.08 0.239 0.22 0.485 0.62 -0.093 -0.11
INDEP -12.52*** -3.80 -16.754*** -5.40 -1.743 -0.96 -14.198*** -2.85 -18.311*** -5.30 -14.597*** -4.21
DUALITY 1.287 1.30 0.326 0.50 0.466 -0.62 0.141 0.12 0.378 0.57 0.915 1.07
COMPETENCE 1.789** 1.80 1.422** 1.90 0.751 -0.80 1.469 1.22 1.551** 1.93 1.576 1.59
OWN 6.325*** 3.05 5.267*** 3.62 1.218 0.73 6.053*** 2.21 6.381*** 3.89 6.386*** 3.22
ROA -5.180 -1.02 3.758 1.03 -10.69** -1.78 -1.842 -0.31 -0.438 -0.12 -3.707 -0.80
LEVERAGE -2.004 -0.81 0.938 0.50 0.061 0.03 0.661 0.21 0.406 0.22 -0.750 -0.33
SIZE 0.179 0.80 0.238 1.32 0.303 1.64 0.339 0.93 0.142 0.80 0.279 1.28
GROWTH -3.546** -1.73 -1.963* -1.66 1.275 1.49 -3.592 -1.27 -2.078** -1.78 -3.587*** -2.22
F (p-value) (0.000) - (0.003) (0.000) (0.011) (0.000)
Adj-R-square 0.7441 0.6162 0.3070 0.6777 0.639 0.7388
Max VIF 1.33 1.31 1.36 1.56 1.32 1.37
Numbers of
observations 250 250 165 85 250 250
Notes: DAcc: Value of discretionary accruals for firm i in year t as measured by the kothari model (2005); SIZE_CA:
Board Size of the firm i in year t, measured by the natural logarithm of the total number of directors; INDEP:
Report of the number of outside directors to the total directors’ number; COMPETENCE: Dummy variable equal
to 1 if a firm has at least one director who has a professional qualification in accounting or finance and 0
otherwise; DUALITY: Dummy variable equal to 1 if the Chairman of the Board of Directors firms i in year t is
also its CEO; OWN: Level of managerial ownership of the firm i in year t; SIZE: The size of firm i in year t,
measured by the natural logarithm of accounting of total assets; ROA : Ratio of income before extraordinary items
to total assets; LEV: The debt level of firm i in year t. Ratio of total debt to total assets; GROWTH: Revenu
Growth = (Revenu t –Revenu t−1)/Revenu t−1.*Significant at 10%; **significant at 5%; ***significant at 1%.

6. Conclusion
The purpose of this study is to examine the effect of discretionary accruals to determine financial
statement fraud, in this regard, we test six models which allow is to know the accruals magnitude on
the probability of determining accounting fraud, also we note, that signs of accruals have an important
effect, the firms engaged more often in a positive discretionary accrual before being engaged in illegal
acts. This is whyour study is in the line with theresearchersof Beasley et al. (2010), Perols and Lougee
(2011) and Trompeter et al. (2013)which showthat the important indicator to detect fraud is
discretionary accruals.
We notice that previous researches have theoretically related the accounting fraud to the
weakness of governance system(Hasnan et al., 2012; Kalbers, 2009; Persons, 2006), this is also found
in our study, the outside directors and ownership have a significant effect on the occurrence of fraud. In
contrast, most previous researchs affirm that the governance mechanisms have no significant effects
such Brown et al. (2011), explain the effect of the regulations and governance characteristics rules are
common in all companies, which makes it more difficult to identify companies engaged in fraudulent
transactions. In addition, they argue that fraudulent companies may change their board of director and
manager. Thus, it's more difficult to identify those which are most likely to engage in fraud. This
corroborates with our study, neither size board of directors nor CEO duality, have a significant effect in
accounting fraud.
This research is the first in France to focus on how discretionary accruals (the absolute and the
different signal) and governance mechanisms are evident in fraudulent companies sanctioned by the
financial market authority (AMF). According to Delvaille et al. (2005), states that the objectives of
public authorities are to improve the functioning of the financial system and the rising level of investor
60 International Research Journal of Finance and Economics - Issue 161 (2017)

confidence. A large stream of research (e.g.Hui et al. 2014, Beneish et al. 2012, Dechow et al.2011,
Jones et al. 2008, Beasley et al. 2010) raise on the question of the fraudulent information detected by
SEC (Securities and Exchange Commission). However, little attention has been given to this
phenomenon in other places of the world like France. Obviously, cases of fraud in the United States
have been widely studied in academic and professional journals.
Among the limitations of this research, in terms of sources of data, we try to identify the
fraudulent firms based on the publications issued by the AMF, but we cannot absolutely guarantee the
absence of a healthy corporate free of fraud. However, this study could have been enriched by
thecomparison between two periods, before and after International Financial Reporting Standards
(IFRS) adoption, and the effect of IFRS in minimizing the rate of financial statement fraud.

References
[1] Agrawala, A. and Chadha, S. ( 2005), "corporate governance and accounting scandals", Journal
of Law and Economics, Vol. 48 No. 2.
[2] Albrecht, G., Krancher, M. and Albrecht, S. (2008), "Asset Misappropriation Resarch White
Paper for the Institute for Fraud Prevention".
[3] Albring, S., Huang, S., Pereira, R. and Xu, X. (2013), "The effects of accounting restatements
on firm growth", Journal of Account. Public Policy Vol. 32 pp. 357–76.
[4] Ball, R. and Shivakumar, L. (2008), "Earnings quality at initial public offerings", Journal of
Accounting and Economics, Vol. 45 No. 2,pp. 324-49.
[5] Beasley, M. (1996), "An empirical analysis of the relation between the board of director
composition and financial statement fraud", The Accounting Review Vol. 71 No. 1,pp. 443–65.
[6] Beasley, M., Branson, B. and Hancock, B. (2010), "Are You Identifying Your Most Significant
Risks Results from a COSO-sponsored survey show that companies need to do a better job in
this area.", COVER STORY.
[7] Beasley, M. and Carcello, J. (2000), " Fraudulent financial reporting: consideration of industry
traits and corporate governance mechanisms", Accounting Horizons, Vol. 14 No. 2,pp. 441-54.
[8] Beneish, M., Lee, C. and Nchols, C. (2012), "Fraud detection and expected returns", Electronic
copy available at: http://ssrn.com/abstract=1998387.
[9] Berle, A., and C. Means. (1932),"The modern corporation and private property. New
Brunswick": NJ: Transaction Publishers.
[10] Brazel, J., Jones, J. and Zimbelman, M. (2009), "Using nonfinancial measures to assess fraud
risk.", Journal of Accounting Research Vol. 47 No. 5,pp. 1135–66.
[11] Brown, P., Dobbie, G. and Jackson, A. (2011), "Measures of the timeliness of earnings", Aust.
Account. Rev, Vol. 21 No. 3,pp. 222–34.
[12] Carcello, J. (2009),"Governance and the common good",Journal of Business EthicsVol.89
No.1, pp.11-18.
[13] Carver, B. (2014), "The retention of directors on the audit committee following an accounting
restatement", J. Account. Public Policy . Vol. 33 pp. 51–68.
[14] Caskey, J. and Hanlon, M. (2013), "Dividend Policy at Firms Accused of Accounting Fraud",
Contemporary Accounting Research, Vol. 30 No. 2,pp. 818–50.
[15] Cassell, C., Myers, J., Myers, L. and Seidel, T. (2014), "Does Auditor Tenure Impact the
Effectiveness of Auditors’ Response to Fraud Risk?", Available at SSRN:
http://ssrn.com/abstract=2448680.
[16] Chen, G., Firth, M., Gao, N. and Rui, O. (2006), "Ownership structure, corporate governance,
and fraud: evidence from China.", Journal of Corporate Finance, Vol. 12 pp. 424-48.
[17] Choi, H., Kim, B. and Lee, J. (2011), "Value relevance of discretionary accruals in the Asian
financial crisis of 1997–1998", J. Account. Public Policy . Vol. 30 pp. 166–87.
International Research Journal of Finance and Economics - Issue 161 (2017) 61

[18] Crutchley, C., Jensen, M. and B, M. (2007), "Climate for Scandal: Corporate Environments that
Contribute to Accounting Fraud", The Financial Review, Vol. 42 pp. 53-73.
[19] Daske, H., L. Hail, C. Leuz, and R. Verdi. (2008). "Mandatory IFRS reporting around the
world: Early evidence on the economic consequences". Journal of accounting researchVol.46
No.5, pp.1085-1142.
[20] Dechow, P. and Dichev, I. (2002), "The quality of accruals and earnings: The role of accrual
estimation errors", The accounting review, Vol. 77 No.1,pp. 35-59.
[21] Dechow, P., Gen, W. and Schrand, C. (2010), "Understanding earnings quality: A review of the
proxies, their determinants and their consequences", Journal of Accounting and Economics Vol.
50, pp. 344-401.
[22] Dechow, P., Larson. C and SLOAN.R (2011), "Predicting Material Accounting Misstatements",
Contempory Accounting Research, Vol. 28 No. 1.
[23] Dechow, P., Sloan, A. and Sweeney, A. (1996), "Causes and consequences of earnings
misstatement: An analysis of firms subject to enforcement actions by the SEC.", Contemporary
Accounting Research, Vol. 13 No. 1,pp. 1–36.
[24] DeFond, M. L., and J. Jiambalvo. (1991),"Incidence and circumstances of accounting
errors",Accounting review, pp. 643-655.
[25] Delvaille, P., G. Ebbers, and C. Saccon. (2005), "International financial reporting convergence:
evidence from three continental European countries",Accounting in europeVol.2 No.1, pp.137-
164.
[26] Erickson, K., M. Hanlon, and E. Maydew. (2006),"Is there a link between executive equity
incentives and accounting fraud",Journal of accounting research,Vol 44 No.1, pp.113-143.
[27] Fama, E. and Jensen, M. (1983), "Agency problems and residual claims.", Journal of law and
Economics,JSTOR., Vol. 26 No. 2.
[28] Fich, E.M. and Shivdasani, A. (2007), "Financial fraud, director reputation, and shareholder
wealth", Journal of Financial Economics, Vol. 86 No. 2,pp. 306-36.
[29] García.Meca, E. and Sánchez.Ballesta, J. (2009), "Corporate Governance and Earnings
Management: A Meta‐Analysis", Corporate Governance: An International Review, Vol. 17 No.
5,pp. 594-610.
[30] Gietzmann, M. and Ireland, J. (2005), "Cost of capital, strategic disclosures and accounting
choice", Journal of Business Finance & Accounting, Vol. 32 No. 3‐4,pp. 599-634.
[31] Gujarati, D. (1988), " Basic econometrics, 2 nd", (New York: McGraw-Hill).
[32] Hasnan, S., Rahman, R.A. and Mahenthiran, S. (2012), "Management motive, weak
governance, earnings management, and fraudulent financial reporting: Malaysian evidence",
Journal of International Accounting Research, Vol. 12 No. 1,pp. 1-27.
[33] Hui, K., Lennox, C. and Zhang, G. (2014), "The Market’s Valuation of Fraudulently Reported
Earning.", Journal of Business Finance & Accounting., Vol. 41 No. 5,pp. 627–51.
[34] Jensen, M. C. (1993), "The modern industrial revolution, exit, and the failure of internal control
systems",the Journal of Finance, Vol.48 No.3, pp.831-880.
[35] Jermias, J. and Gani, L. (2014), "The impact of board capital and board characteristics on firm
performance", The British Accounting Review Vol. 46 pp. 135–53.
[36] Jia, C., Ding, S., Li, L. and Wu, Z. (2009), "Fraud, Enforcement Action, and the Role of
Corporate Governance: Evidence from China", Journal of Business Ethics Vol. 90 pp. 561–76.
[37] Jones, J. (1995), "Earnings Management During Import Relief Investigations.", Journal of
Accounting Research Symposium., Vol. 29 No. 2,pp. 193-228.
[38] Jones, K., Krishnan, G. and Melendrez, K. (2008), "Do Models of Discretionary Accruals
Detect Actual Cases of Fraudulent and Restated Earnings? An Empirical Analysis", Journal of
Contemporary Accounting & Economics-Wiley Online Library.
[39] Karjalainen, J. (2011), Audit quality and cost of debt capital for private firms: Evidence from
Finland. International Journal of Auditing, 15(1), 88-108.
62 International Research Journal of Finance and Economics - Issue 161 (2017)

[40] Kalbers, L. (2009), "Fraudulent financial reporting, corporate governance and ethics: 1987-
2007.", Review of Accounting and Finance, Vol. 8 No. 2,pp. 187 - 209.
[41] Kamarudin, K., W. Ismail, and W. Mustapha. (2012),"Aggressive financial reporting and
corporate fraud", Procedia-Social and Behavioral Sciences,Vol. 65, pp.638-643.
[42] Kothari, S., Leone, A. and Charles, W. (2005), "Performance matched discretionary accrual
measures.", Journal of Accounting and Economics Vol. 39 No. 1,pp. 163-97.
[43] Lennox, C., Lisowsky, P. and Pittman, J. (2013), "Tax Aggressiveness and Accounting Fraud.",
Journal of Accounting Research.Printed in U.S.A., Vol. 00 No. 0.
[44] Lim, M., How, J. and Verhoeven, P. (2014), "Corporate ownership, corporate governance
reform and timeliness of earnings: Malaysian evidence.", Journal of Contemporary Accounting
& Economics 10, pp. 32-45.
[45] Lisic, L., Silveri, S., Song, Y. and Wang, K. (2015), "Accounting fraud, auditing, and the role
of government sanctions in China", Journal of Business Research, Vol. 68 No. 6,pp. 1186-95.
[46] Myers, R. (1990), "Classical and Modern Regression with Application, DUXBURY", Pacific
Grove, CA, Vol. 93950.
[47] Park, Y. and Shin, H. (2004), "Board composition and earnings management in Canada",
Journal of Corporate Finance, Vol. 10 pp. 431 – 57.
[48] Perols, J. and Lougee, B. (2011), "The relation between earnings management andfinancial
statement fraud", Advances in Accounting, incorporating Advances in International
Accounting, Vol. 27 pp. 39–53.
[49] Persons, O. (2006), "The Effects of Fraud and Lawsuit Revelation on U.S. Executive Turnover
and Compensation.", Journal of Business Ethics. , Vol. 64 pp. 405–19.
[50] Rezaee, Z. (2005),"Causes, consequences, and deterence of financial statement fraud", Critical
Perspectives on Accounting, Vol 16, pp. 277-298.
[51] Skousen, C.J. and Wright, C.J. (2006), "Contemporaneous risk factors and the prediction of
financial statement fraud", Available at SSRN 938736.
[52] Stolowy, H. and Breton, G. (2003), "La gestion des données comptables: une revue de la
littérature.", Comptabilité contrôle audit.
[53] Trompeter, G., Carpenter, T., Desai, N., Jones, K. and Riley, R. (2013), "A Synthesis of Fraud-
Related Research.", AUDITING: A Journal of Practice & Theory Vol. 32 No. 1,pp. 287-321.
[54] Velikonja, U. (2011). "Leverage, Sanctions, and Deterrence of Accounting Fraud". UC Davis
Law Review,Vol.44, pp.1281-1345.
[55] Wakil, G., and P. Alam. (2012),"Conservatism, Accrual Persistence Asymmetry, and Stock
Returns". Paper read at CAAA Annual Conference
[56] Warfield, T., J. Wild, and K. Wild. (1995),"Managerial ownership, accounting choices, and
informativeness of earnings",Journal of accounting and economicsVol.20 No. 1, pp. 61-91.
[57] Watts, R.L. and Zimmerman, J.L. (1990), "Positive accounting theory: a ten year perspective",
Accounting review, pp. 131-56.
[58] Wuerges, A. and Borba, J. (2010), "Accounting fraud detection : is it possible to quantify
undiscovered cases?", Social Science Research Network.
[59] Xie, B., Davidson, W., DaDalt, P. and Abbott, L. (2003), "Earnings management and corporate
governance: the role of the board and the audit committee", Journal of Corporate Finance Vol.
9 pp. 295 – 316.
[60] Xie, H. (2001), "The mispricing of abnormal accruals", The accounting review, Vol. 76 No.
3,pp. 357-73.

View publication stats


ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

PENGARUH AUDIT TENURE, AUDIT FEE, ROTASI


AUDITOR, SPESIALSIASI AUDITOR, DAN UMUR
PUBLIKASI PADA KUALITAS AUDIT

I Gusti Ngurah Indra Pramaswaradana1


Ida Bagus Putra Astika2
1
Fakultas Ekonomi dan Bisnis Universitas Udayana (Unud) Bali, Indonesia
email: turahindra95@yahoo.com / telp: +62 812 389 07837
2
Fakultas Ekonomi dan Bisnis Universitas Udayana (Unud) Bali, Indonesia

ABSTRAK
Penelitian ini bertujuan memberikan bukti empiris tentang pengaruh audit tenure,
audit fee, rotasi auditor, spesialisasi auditor, dan umur publikasi pada kualitas audit.
Populasi penelitian ini adalah seluruh perusahaan manufaktur yang terdaftar di Bursa
Efek Indonesia periode 2013-2015. Metode penentuan sampel menggunakan metode
purposive sampling sehinggamemperoleh 67 sampel perusahaan. Metode
pengumpulan data menggunakan metode observasi non partisipan dengan cara
mengunduh data dari situs resmi bursa efek Indonesia.Analisis data menggunakan
teknik regresi logistik. Hasil uji hipotesis adalah audit tenure berpengaruh negatif
pada kualitas audit, audit fee berpengaruh positif pada kualitas audit, sedangkan
rotasi, spesialisasi, serta umur tidak berpengaruh pada kualitas audit
Kata Kunci : audit tenure, audit fee, rotasi auditor, spesialisasi auditor, umur
publikasi, kualitas audit.

ABSTRACT
This study aim to provide empirical evidence about the effect of audit tenure, audit
fee, auditor rotation, auditor specialization, and age of publication on audit quality.
The population in this study is all of manufacturing company listed in Indonesia
Stock Exchange on period 2013-2015. The sampling method used the method of
purposive sampling with the result 67 company sample. Data collection method used
the method of non participant observation with download data from official webside
Indonesia stock exchange. Data analysis used the technique of logistic regression.
The result of hypothesis is audit tenure had negative effect on audit quality, audit fee
had positive effect on audit quality, while rotation, specialization, and age of
publication had no effect on audit quality.
Key Word : audit tenure, audit fee, auditor rotation, auditor specialization, age of
publication, audit quality.

PENDAHULUAN

Entitas yang sedang tumbuh dan berkembang dalam bidang bisnis. Perkembangan

bisnis dapat terlihat dari laporan keuangan. Manajer perusahaan pasti membuat

168
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

laporan keuangan. Laporan ini bermanfaat bagi pengguna laporan seperti investor

maupun kreditor untuk bisa mendapatkan informasi keuangan perusahaan. Pengguna

laporan perlu mengecek informasi yang terdapat dalam laporan keuangan itu tidak

mengandung unsur kesalahan dengan menggunakan jasa auditor independen. Jasa

auditor independen bertugas melakukan audit dan memastikan informasi dari laporan

keuangan tidak mengandung unsur kesalahan sehingga nantinya dapat dijadikan

pedoman dalam pengambilan keputusan. Audit merupakan suatu kegiatan

mengumpulkan bukti tentang informasi dalam laporan keuangan kemudian

menelusuri dan mengungkapkan informasi yang sebenarnya. Menurut Enofe et al.,

(2013) audit merupakan pemeriksaan laporan keuangan perusahaan yang dilakukan

secara independen. Jensen dan Meckling (1967) menyatakan audit merupakan proses

untuk mengawasi pihak manajer dalam membuat laporan keuangan dan memberikan

hasil yang sesuai kepada pemegang saham.

Syarat informasi yang ada dalam laporan keuangan itu harus reliabel dan

relevan (Singgih dan Bawono, 2010). Dalam era sekarang, semakin banyak yang

membutuhkan jasa auditor independen untuk melakukan audit laporan keuangan

perusahaan dan ini menjadikan kebutuhan pokok sebelum melakukan pengambilan

keputusan berbisnis (Baotham, 2007). Para pengguna laporan keuangan tidak bisa

menilai kebenaran informasi yang ada dalam laporan keuangan jika tidak

menggunakan jasa auditor independen.

Seorang auditor memerlukan pengetahuan khusus seperti ilmu akuntasi,

pengetahuan dalam audit, dan industri klien untuk bisa melaksanakan tugasnya

169
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

dalam melakukan audit. Ada tiga standar audit yang dapat dijadikan pedoman dalam

mengaudit laporan keuangan yaitu standar umum, standar pekerjaan lapangan, dan

standar pelaporan (IAPI, 2011). Standar audit ini sangat berkaitan dengan beberapa

aspek penting seperti perencanaan, pengawasan, pendeteksi kecurangan, penentuan

resiko, peningkatan kredibilitas, dan pemberian kepuasan klien (Abdelrahman,

2014). Dengan adanya standar audit yang berlaku dapat membuat auditor lebih

bertanggung jawab sehingga dapat meningkatkan keyakinan klien terhadap auditor.

Beberapa tahun silam, pernah terjadi peristiwa atau skandal yang melibatkan

auditor ternama. Perusahaan Enron di Amerika Serikat mengalami kebangkrutan

akibat ulah jasa auditor yang digunakan yaitu jasa Kantor Akuntan Publik (KAP)

milik Arthur Andersen. Perusahaan ini bangkrut disebabkan karena adanya masalah

independensi auditor terhadap klien. Selama 20 tahun Enron telah menggunakan jasa

auditor Arthur Andersen, hal ini membuat hubungan antara Enron dan Arthur

semakin dekat dan akrab melebihi teman. Akibat dari kedekatannya, Arthur menjadi

segan dan tidak berani mengungkapkan jika terdapat salah saji dalam laporan

keuangannya. Peran dan kualitas seorang auditor dalam mengaudit sempat menjadi

topik pembicaraan saat itu dan mendapat kritikan yang banyak, sehingga membuat

masyarakat waspada dalam menggunakan jasa audit laporan keuangan.

Kualitas audit merupakan probabilitas seorang auditor dalam mendeteksi dan

mengungkapkan apabila terdapat pelanggaran dalam sistem akuntansi kliennya

(DeAngelo, 1981). Seorang auditor dapat dikatakan berkualitas jika mampu

menemukan dan berani mengungkapkan adanya unsur kesalahan dalam laporan

170
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

keuangan. Edosa et al., (2013) menyatakan bahwa kualitas yang dimiliki auditor

dapat dilihat jika dapat ditemukan adanya kesalahan saji dalam laporan keuangan

perusahaan. Kantor akuntan publik yang besar atau KAP Big 4 dianggap memiliki

pengetahuan, pengalaman, dan reputasi yang lebih baik dibandingkan dengan KAP

non Big 4 atau kecil sehingga KAP Big 4 dianggap mampu menghasilkan kualitas

audit yang lebih baik dibandingkan KAP non Big 4 (Kurniasih dan Rohman, 2014).

Kualitas audit berkaitan dengan beberapa faktor seperti audit tenure, audit fee, rotasi

auditor, spesialisasi auditor, dan umur publikasi.

Audit tenure atau masa perikatan audit dapat mempengaruhi kualitas audit.

Masa perikatan audit sempat menjadi perdebatan publik. Apabila masa perikatan

yang dilakukan dalam waktu lama akan membuat kedekatan yang berlebihan antara

klien dan auditor sehingga independensi dan objektivitas auditor nantinya akan

menurun. Pernyataan tersebut serupa dengan penelitian Al-Thuneibat et al., (2011)

yang menyatakan masa perikatan cukup lama dapat mempengaruhi objektivitas

auditor dalam bekerja. Jika dilakukan dalam waktu singkat akan membuat auditor

kurang maksimal dalammemperoleh informasi dan memahami lingkungan

perusahaan klien. Hamid (2013) menyatakan bahwa masa perikatan yang singkat

membuat auditor kekurangan waktu untuk memahami lingkungan bisnis klien

sehingga kurang mampu menilai informasi secara lebih baik.

Pemberian imbalan jasa atau fee sesuai keinginan dapat memotivasi auditor

untuk meningkatkan kinerjanya dalam mengaudit laporan perusahaan (Maharani,

2014). Jika semakin besar imbalan yang diberikan klien kepada auditor dapat

171
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

membuat auditor memaksimalkan kemampuannya dalam melakukan audit dan

memperbaiki kinerjanya menjadi lebih baik. Jika imbalan yang diberikan kecil dapat

membuat auditor menjadi tidak termotivasi untuk meningkatkan kemampuan dan

melaksanakan tugas dengan baik. Besarnya imbalan yang akan diberikan klien

kepada auditor tergantung dari kemampuan dan keahlian yang dimiliki auditor, serta

tarif jasa yang dikenakan oleh auditor tersebut (Hanjani, 2014).

Adanya aturan dari pemerintah tentang rotasi auditor dapat berdampak baik

bagi klien. Myers (2013) menyatakan tujuan adanya rotasi auditor dalam mengaudit

perusahaan untuk mencegah klien dan auditor memiliki kedekatan yang berlebihan.

Apabila perusahaan klien semakin cepat melakukan rotasi maka akan dapat

mengurangi dan mencegah klien berhubungan dekat dengan auditor tersebut

sehingga auditor pun tidak akan mengalami masalah independensi. Namun jika

perusahaan tidak pernah melakukan rotasi maka akan terjalin kedekatan yang sangat

erat antara kedua belah pihak sehingga dapat berpengaruh terhadap objektivitas

auditor. Penelitian yang dilakukan oleh Velte dan Stiglbauer (2012) serta Siregar et

al., (2012) menyatakan rotasi akuntan publik atau auditor diharapkan dapat

menciptakan suasana baru dan mampu bertidak lebih objektif.

Pengalaman yang dimiliki auditor dalam melakukan audit perusahaan yang

sama menjadikan auditor tersebut memiliki spesialisasi. Auditor spesialis memiliki

pengalaman audit perusahaan yang sama sebelumnya dan lebih memahami kondisi

perusahaan. Klien yang menggunakan jasa auditor spesialis dianggap mampu lebih

cepat dan baik dalam mendeteksi dan menemukan kesalahan yang terdapat pada

172
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

laporan keuangan yang diaudit. Klien yang tidak menggunakan jasa auditor spesialis

dianggap kurang mampu melakukan audit secara lebih baik dan cepat. Spesialisasi

auditor pada suatu industri membuat auditor memiliki suatu pengalaman dan

pengetahuan yang cukup tentang kondisi perusahaan (Fitriany, 2011). Auditor

spesialis diyakini memiliki kualitas laporan yang baik dan mampu mendeteksi

informasi secara akurat (Romanus, 2008).

Semakin lama umur perusahaan dipublikasikan, perusahaan tersebut

dianggap memliki kualitas dan manajemen yang bagus. Manajemen dianggap

mampu mengelola keuangan secara efektif dan efisien serta mampu menyediakan

informasi keuangan dengan benar. Hal ini membuat auditor mudah dalam

mendapatkan informasi dan melakukan audit. Manajer yang berkualitas adalah

manajer yang memiliki pengalaman dalam bidang pengendalian internal perusahaan

(Fernando, 2010). Semakin singkat umur publikasi suatu perusahaan, maka

perusahaan tersebut dianggap memiliki kualitas dan manajemen yang kurang bagus.

Manajer dianggap tidak mampu menyediakan informasi dengan benar dan

menghasilkan kualitas laporan yang kurang akurat bagi pengguna informasi.

Penelitian ini berfokus pada kualitas audit perusahaan manufaktur. Penelitian

ini dibuat berdasarkan acuan dari penelitian yang dilakukan Maharani tahun 2014.

Perbedaannya terletak pada pergantian beberapa variabel, tahun pengamatan, dan

proksi, serta adanya penambahan variabel bebas. Penjelasan singkat dari latar

belakang dapat merumuskan masalah yang terdapat dalam penelitian ini yaitu

bagaimana pengaruh audit tenure, audit fee, rotasi auditor, spesialisasi auditor, dan

173
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

umur publikasi pada kualitas audit di perusahaan manufaktur tahun 2013-

2015.Tujuan dari penelitian ini adalah untuk memberikan bukti empiris pengaruh

audit tenure, audit fee, rotasi auditor, spesialisasi auditor, dan umur publikasi ada

kualitas audit di perusahaan manufaktur tahun 2013-2015.Kegunaan dari penelitian

ini adalah diharapkan dapat memberikan informasi bagi pihak yang berkepentingan

sebagai bahan referensi untuk melakukan penelitian dan riset selanjutnya.

Adanya motivasi dapat mendorong seseorang untuk melakukan segala

sesuatu. Motivasi tersebut bisa berasal dari dalam dan luar. Teori ini berkaitan

dengan audit fee. Auditor harus memiliki motivasi yang jelas dalam melasanakan

tugasnya agar dapat mencapai tujuan audit dan tujuan perusahaan. Ada beberapa

faktor yang dapat memotivasi auditor meningkatkan kemampuannya dalam bekerja

seperti adanya imbalan yang besar dan pemberian hadiah atau penghargaan (Tan,

2000). Auditor yang termotivasi untuk mendapatkan imbalan yang besar dari klien

akan berusaha menujukkan kemampuan terbaik yang dimilikinya serta menghasilkan

kualitas yang baik.

Taat dan patuh pada sesuatu hal seperti aturan yang berlaku, menunujukkan

seseorang memiliki sikap patuh. Masyarakat dari berbagai kalangan pun harus

mempunyai sikap taat dan patuh pada aturan yang berlaku terlebih dari aturan hukum

di Indonesia. Auditor juga dituntut harus patuh pada peraturan yang dibuat oleh

menteri keuanganNo.17 tahun 2008 mengenai jasa akuntan publik. Isi dari aturan

tersebut adalah ada batasan waktu selama tiga tahun untuk akuntan publiknya dalam

memberi jasa audit sedangkan untuk kantor akuntan publiknya selama lima tahun

174
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

berturut-turut. Aturan seperti ini dibuat dengan maksud mencegah klien agar tidak

memiliki hubungan spesial dengan auditor sehingga objektivitas kerja auditor tidak

akan berpengaruh.Selain peraturan tersebut, auditor harus mematuhi pedoman

standar audit yang dibuat oleh Institut Akuntan Publik Indonesia (IAPI). Pedoman

tersebut bertujuan membantu auditor agar dapat menjalankan prosedur audit dengan

benar.

Belajar merupakan salah satu cara yang dapat dilakukan untuk

mengembangkan dan menambah wawasan ke arah yang lebih baik. Proses

pembelajaran tersebut berhubungan dengan proses menambah pengetahuan dan

pengalaman seseorang. Auditor yang telah lama mengaudit di suatu perusahaan pasti

akan lebih memahami kondisi lingkungan perusahaannya dan mampu dengan mudah

untuk melaksanakan prosedur audit dibandingkan dengan auditor yang belum pernah

mengaudit. Hal ini disebabkan karena auditor lama memiliki lebih banyak

pengalaman dan pemahaman yang cukup dibandingkan auditor baru. Pengalaman

yang dimiliki auditor dapat menjadikannya mempunyai spesialisasi dalam mengaudit

laporan keuangan pada perusahaan tersebut karena dapat melakukan proses audit

secara efektif dan efisien sehingga dapat menghasilkan kualitas yang lebih baik.

Proses belajar juga dilakukan pada perusahaan untuk meningkatkan

kualitasnya. Semakin lama perusahaan dipublikasikan menandakan perusahaan

tersebut dapat bertahan dan dianggap memiliki kualitas serta manajemen yang bagus.

Manaejemen dalam hal ini dianggap mampu menyediakan sumber informasi

keuangan dengan akurat sehingga memudahkan auditor untuk menemukan dan

175
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

melaporkan adanya salah saji yang terdapat dalam laporan keuangaan perusahaan.

Kualitas audit yang dihasilkan auditor berhubungan dengan kemampuannya dalam

menemukan adanya kecurangan dalam laporan keuangan kemudian berani

mengungkapnya. Pengguna laporan melihat kualitas yang dihasilkan auditor dari segi

kemampuannya mendeteksi adanya unsur kecurangan. Jika auditor bekerja sesuai

standar yang berlaku, mampu menilai resiko yang terjadi dan menjaga reputasi maka

auditor tersebut dikatakan berkualitas.

Tenur audit merupakan lamanya perikatan kantor akuntan publik yang terjadi

dengan perusahaan klien (Maharani, 2014). Masa perikatan sempat menjadi

perdebatan publik karena jika dilakukan terlalu lama maupun singkat akan

berdampak buruk bagi auditor dan klien. Jika telalu lama dapat menyebabkan

kedekatan yang dapat berpengaruh pada independensi auditor dalam melaporkan

adana kesalahan. Namun jika perikatannya singkat dapat membuat auditor tidak

memiliki informasi yang cukup untuk mempelajari kondisi perusahaan terlebih

dahulu. Pemberian fee merupakan imbalan jasa yang diberikan kepada auditor karena

telah bekerja untuk mengaudit di perusahaan klien. Imbalan tersebut dapat diberikan

klien jika auditor telah selesai melakukan tugasnya. Alasan lain diberikannya

imbalan untuk membuat auditor menjadi termotivasi meningkatkan kinerjanya lebih

giat sehingga akan mendapatkan hasil yang terbaik. Sebelum memberikan imbalan

jasa, antara auditor dan klien sudah membuat kesepakatan pembayaran terlebih

dahulu.

176
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

Pergantian jasa auditor dalam mengaudit suatu perusahaan sangat penting

dilakukan karena untuk mengurangi kedetakan yang terjalin antara kedua belah

pihak. Rotasi berkaitan dengan adanya batasan waktu untuk akuntan publik dalam

bekerja memberikan jasa audit pada suatu perusahaan. Seperti yang tercantum dalam

peraturan menteri keuangan No. 17 tahun 2008 bahwa akuntan publik dibatasi

memberikan jasa audit selama tiga tahun sedangkan untuk kantor akuntan publik

dibatasi selama lima tahun berturut-turut. Pergantian auditor dapat terjadi secara

sukarela maupun wajib. Pengalaman audit perusahaan yang dimiliki auditor

membuatnya memiliki kemampuan khusus atau memiliki spesialisasi perusahaan

tertentu. Adanya spesialisasi menyebabkan auditor mendeteksi adanya salah saji

lebih baik, meningkatkan efisiensi, dan mampu menilai keandalan dalam

mengungkapkan kesalahan yang terdapat pada laporan keuangan. Banyak

masyarakat yang ingin menggunakan auditor spesialis untuk menilai apakah

informasi keuangan yang dibuat manajer perusahaan itu akurat atau tidak. Penelitian

ini berfokus pada spesialis auditor pada lingkup industri manufaktur.

Kemampuan perusahaan untuk mampu bertahan, bersaing, dan memnfaatkan

peluang bisnis dalam perkenomian ditunjukkan melalui umur publikasinya. Lamanya

perusahaan yang terdaftar atau dipublikasikan di Bursa Efek Indonesia dianggap

memiliki kualitas yang baik baik dari segi manajemen maupun operasionalnya. Pihak

manajemen perusahan mampu menyediakan informasi dengan jelas. Hal tersebut

menguntungkan bagi auditor karena akan mudah mendapatkan sumber informasi dan

mampu melakukan audit laporan dengan cepat.

177
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

Maharani (2014) melakukan penelitian tentang kualitas audit. Variabel bebas

yang digunakan adalah audit fee, audit tenure, rotasi audit, dan spesialisasi audit

sedangkan untuk variabel terikatnya menggunakan kualitas audit. Populasi yang

digunakan adalah perusahaan manufaktur di bursa efek Indonesia periode 2013-

2015. Teknik analisisnya menggunakan regresi logistik dengan hasil yang

menyatakan audit tenure, rotasi audit, dan spesialisasi audit tidak berpengaruh pada

kualitas audit sedangkan audit fee berpengaruh pada kualitas audit. Penelitian yang

dilakukan Kurniasih dan Rohman (2014) membahas tentang kualitas yang dihasilkan

auditor pada perusahaan manufaktur di bursa Efek periode 2008-2012. Rotasi audit,

audit fee, audit tenure sebagai variabel independen dan kualitas audit sebagai

variabel dependennya. Pengujian hipotesis menggunakan analisis regresi logistik.

Hasilya adalah audit fee dan rotasi audit bepengaruh postif sedangkan audit tenure

berpenaruh negatif pada kualitas audit.

Panjaitan (2014) meneliti pengaruh audit tenure, ukuran kap, dan spesialisasi

auditor terhadap kualitas audit. Sampel yang digunakan adalah perusahaan

manufaktur di bursa efek Indonesia tahun 2010-2012. Teknik analisis yang

digunakan adalah regresi linier berganda dengan hasilnya audit tenure berpengaruh

negatif, spesialisasi berpengaruh positif, dan ukuran kap tidak berpengaruh pada

kualitas audit.

Hartadi (2012) juga meneliti tentang kualitas audit. Variabel bebasnya yaitu

fee audit, rotasi kap, dan reputasi audit sedangkan variabel terikatnya adalah kualitas

audit. Alat analisisnya adalah regresi logistik. Hasil uji hipotesisnya menyatakan fee

178
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

audit berpengaruh signifikan terhadap kualitas audit namun rotasi dan reputasi audit

tidak berpengaruh signifikan. Penelitian Permana (2012) yang menjelaskan

bagaimana pengaruh masa perikatan audit serta ukuran kap terhadap kualitas audit

dengan sampel yang digunakan adalah perusahaan manufaktur. Teknik analisisnya

menggunakan regresi logistik. Hasil dari penelitian ini menyatakan kedua variabel

tersebut tidak berpengaruh terhadap kualitas audit. Semakin lama masa perikatan

auditor dengan klien, maka akan berdampak buruk bagi kualitas auditor tersebut.

Adanya kedekatan secara mendalam antara klien dan auditor dapat membuat rasa

objektif yang dimiliki auditor tersebut berkurang dan merasa takut untuk

mengungkap keadaan perusahaan sebenarnya. Objektivitas jasa auditor sangat

penting dan dapat menjadi ancaman ketika melakukan perikatan yang cukup lama

dengan auditor yang bersangkutan. Pernyataan tersebut didukung oleh penelitian

Chen et al., (2004) serta Giri (2010) yang membuktikan bahwa masa perikatan

audititu berpengaruh negatif dan signifikan terhadap kualitas audit. Hipotesis yang

dapat diajukan dari penjelasan di atas yaitu:

H1 : Audit tenure berpengaruh negatif pada kualitas audit.

Pembayaran imbalan jasa kepada auditor dilakukan setelah selesai melakukan

tugasnya. Imbalan yang diberikan berdasarkan atas perjanjian yang telah dibuat

sebelumnya. Semakin besar upah yang diberikan, maka auditor semakin terdorong

untuk meningkatkan kemampuan dan kinerjanya sehingga mendatangkan hasil yang

baik. Penelitian dari Kurniasih dan Rohman (2014) menyatakan bank serta

perusahaan besar yang ada di Lebanon berani memberikan upah yang besar kepada

179
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

auditor demi menjamin informasi yang diperoleh nanti hasilnya akurat sehingga

secara tidak langsung dapat juga meningkatkan kredibilitas laporan tahunan dari

perusahaan tersebut. Hipotesis yang dapat diajukan dari penjelasan di atas yaitu:

H2 : Audit fee berpengaruh positif pada kualitas audit

Kebijakan rotasi auditor yang dibuat pemerintah sangat bagus untuk

mencegah dan mengurangi kedekatan spesial yang terjalin dengan auditor. Suatu

perusahaan harus pernah melakukan pergantian auditor dan menggunakan jasa

akuntan publik lain demi terciptanya suasana yang lebih baru sehingga auditor dapat

lebih bersikap objektif. Jika perusahaan tidak pernah mengganti jasa auditornya,

maka akan dapat mengancam independensi seorang auditor tersebut. Semakin cepat

perusahaan klien melakukan pergantian auditor, diharapkan akan menghasilkan

kualitas audit yang lebih bagus karena dapat mencegah auditor dan klien memiliki

kedekatan khusus yang berpengaruh pada independensi. Menurut Velte dan

Stiglbauer (2012) dengan melakukan rotasi audit, makaauditor akandapat

meningkatkan kualitasnya. Hipotesis yang dapat diajukan dari penjelasan di atas

yaitu:

H3 : Rotasi auditor berpengaruh positif pada kualitas audit

Kebutuhan mengenai auditor spesialis sangat banyak terutama di bidang

industri manufaktur. Auditor yang memiliki spesialisasi dianggap dapat mendeteksi

kesalahan lebih baik, cepat, dan efisien. Klien juga menerima informasi yang

relevan, akurat dan andal dari laporan keuangan diaudit. Oleh karena itu, jasa auditor

spesialis banyak digunakan untuk mengaudit laporan perusahaan klien. Semakin

180
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

sering auditor mengaudit suatu perusahaan diharapkan semakin bertambah

kemampuan dan pengalaman yang dimiliki auditor sehingga dianggap mampu

menghasilkan kualitas audit yang baik. Auditor spesialis industri tertentu memiliki

kualitas lebih bagus daripada yang tidak memiliki spesialisasi (Novianti, 2012).

Hipotesis yang dapat diajujan dari penjelasan di atas yaitu:

H4 : Spesialisasi auditor berpengaruh positif pada kualitas audit

Perusahaan yang sudah lama terdaftar di bursa efek menunjukkan perusahaan

tersebut masih ada dan mampu bertahan. Semakin lama umur perusahaan yang

dipublikasikan dianggap memiliki kualitas yang bagus dalam berbagai hal seperti

manajemen dan operasionalnya. Manajer dianggap mampu menyediakan sumber

informasi yang dengan jelas. Hal ini dapat memudahkan auditor dalam melakukan

audit karena tersediannya informasi untuk mendeteksi jika terjadi kecurangan.

Hidayat dan Elizabet (2007) serta Bestivano (2013) menyatakan pendapat yang

serupa, semakin lama umur publikasi dapat mengurangi tindakan kejahatan yang

dilakukan manajemen karena rendahnya tingkat akrual diskresioner perusahaan. Hal

ini membuat auditor menjadi lebih cepat dalam menemukan dan melaporkan adanya

kesalahan pada laporan tersebut. Hipotesis yang dapat diajukan dari penjelasan di

atas yaitu:

H5 : Umur publikasi berpengaruh positif pada kualitas audit.

METODE PENELITIAN

Lokasi penelitian ini adalah di Bursa Efek Indonesia kemudian masuk ke situs

resminya. Penelitian ini merupakan penelitian kuantitatif yang menggunakan data

181
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

dalam bentuk angka lalu menggunakan alat statistik untuk menganalisisnya. Variabel

yang digunakan yaitu variabel bebas dan terikat. Variabel bebas berupa audit tenure,

audit fee, rotasi auditor, spesialisasi auditor, dan umur publikasi sedangkan variabel

terikatnya berupa kualitas audit. Audit tenure adalah lamanya masa perikatan auditor

dengan perusahaan klien. Cara mengukurnya dengan menghitung lamanya waktu

auditor dan klien melakukan perikatan (Sinaga, 2012).

Audit fee adalah pendapatan yang diberikan klien kepada auditor sebagai

upah jasa melakukan audit. Cara mengukurnya menggunakan proksi jasa

professional yang ada dalam laporan keuangan kemudian dihitung menggunakan

logaritma natural (Kurniasih dan Rohman, 2014). Rotasi auditor adalah pergantian

auditor yang ada dalam kantor akuntan publik untuk memberikan jasa audit kepada

klien. Rotasi diukur menggunakan variabel dummy yang bernilai 1 jika perusahaan

klien melakukanpergantian auditor dan bernilai 0 jika tidak melakukan pergantian

(Perdana, 2014)

Spesialisasi auditor adalah kemampuan yang didapatkan berdasarkan

pengalaman mengaudit perusahaan sebelumnya. Spesialisasi diukur menggunakan

variabel dummy yang bernilai 1 jika memakai jasa spesialis dan bernilai 0 jika tidak

memakai jasa spesialis. Auditor yang sudah pernah mengaudit dapat dikatakan

memiliki pengalaman (Balsam et al., 2003) dalam (Herusatya, 2009). Umur

publikasi adalah usia awal tercatat dan terdaftarnya suatu entitas sampai dapat

bertahan. Entitas tersebut telah terdaftar di bursa resmi. Cara mengukurnya dengan

182
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

menghitung usia entitas sejak awal terdaftar sampai tahun yang dijadikan

pengamatan.

Kualitas audit adalah keahlian yang dimiliki auditor untuk bisa mendeteksi

dan berani mengungkapkan jika terdapat kesalahan yang ada pada laporan keuangan

entitas. Kualitas ini diproksikan dengan ukuran kap kemudian diukur menggunkan

dummy yaitu bernilai 1 jika kap big 4 yang mengaudit laporan dan bernilai 0 jika

tidak memakai kap big 4 dalam mengaudit laporan. Jenis data yang digunakan adalah

data kuantitatif. Data ini berupa laporan keuangan audit dan entitas dari beberapa

sampel perusahaan manufaktur di bursa efek Indonesia periode 2013-2015. Sumber

data yang didapatkan berasal dari data sekunder yaitu profil-profil perusahaan

manufaktur yang ada di bursa efek Indonesia selama tahun 2013-2015.

Populasi penelitiannya adalah seluruh perusahaan manufaktur yang tercatat di

bursa efek Indonesia selama periode 2013-2015. Sampel yang digunakan hanya

beberapa dari keseluruhan jumlah perusahaan yang ada. Dalam menentukan sampel,

metode yang digunakan adalah purposive sampling yang berisi kriteria tertentu. Ada

beberapa kriteria perusahaan yang harus dipenuhi untuk dijadikan sampel yaitu: 1)

Perusahaan tersebut harus tercatat di bursa efek Indonesia selama peride yang

dijadikan pengamatan yaitu 2013-2015. 2) Mampu mempublikasikan laporan

keuangan yang sudah diaudit pada akhir tahun selama periode pengamatan. 3)

Mampu menyediakan data-data lengkap yang berkaitan dengan variabel penelitian

yang digunakan

183
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

Metode yang digunakan untuk mengumpulkan data adalah metode observasi

non partisipan. Peneliti tidak harus terjun langsung ke lapangan untuk mendapatkan

data. Teknik analisisnya menggunakan regresi logistik karena menggunakan variabel

dummy pada variabel terikatnya. Ada beberapa langkah yang harus dilakukan

sebelum melakukan analisis data yaitu menilai kelayakan model, menilai

keseluruhan model, menguji koefisien determinasi, menguji multikolinearitas,

menguji matriks klasifikasi, dan mendapatkan model regresi yang terbentuk

(Ghozali, 2012). Model regresi logistik yang digunakan adalah sebagai berikut:
𝐾𝐴
Ln (1−𝐾𝐴 ) = α + β1Tenur + β2Fee + β3Rotasi + β4Spesialisasi + β5Umur + e………(1)

Keterangan:
𝐾𝐴
Ln (1−𝐾𝐴 ) = Kualitas audit yang diukur menggunakan proksi ukuran kap

dihitung menggunakan variabel dummy yang bernilai 1 jika


diaudit oleh kap big 4 dan bernilai 0 jika tidak diaudit oleh
kap big 4.
α = Konstanta
β(12345) = Koefisien model regresi (X12345)
Tenur = Diukur dengan menghitung lamanya klien dan auditor
melakukan perikatan dalam memberi jasa audit.
Fee = Diukur dengan menggunakan proksi jasa profesional yang
ada pada catatan atas laporan keuangan kemudian dihitung
menggunakan logaritma natural.
Rotasi = Diukur menggunakan variabel dummy. Bernilai 1 jika
perusahaan melakukan pergantian auditor dan 0 jika tidak
melakukan pergantian.

184
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

Spesialisasi = Diukur menggunakan variabel dummy. Bernilai 1 jika klien


menggunakan auditor spesialis dan 0 jika tidak
menggunakan auditor spesialis.
Umur = Diukur dengan menghitung usia dari awal tercatatnya
perusahaan dalam Bursa Efek Indonesia sampai periode
pengamatan yang dilakukan.
e = Standar eror

HASIL DAN PEMBAHASAN

Tabel 1.
Hasil Uji Statsistik Deskripsi
N Minimum Maximum Mean Std. Deviation
Tenur 201 1 3 1.87 0.82
Fee 201 15.24 28.29 21.66 2.09
Rotasi 201 0 1 0.09 0.29
Spesialisasi 201 0 1 0.79 0.41
Umur 201 1 33 19.49 7.57
Kualitas 201 0 1 0.46 0.50
Sumber: data diolah, 2016

Tabel 1 menjelaskan hasil dari deskripsi statistik penelitian yang berkaitan

dengan beberapa variabel antara lain: Tenur memiliki nilai maksimum 3 yang berarti

masa perikatan yang paling lama dilakukan adalah 3 tahun. Nilai minimumnya 1

yang berarti masa perikatan paling cepat adalah 1 tahun. Rata-rata dari jumlah

seluruh perusahaan sampel melakukan perikatan selama 1,87 tahun dengan standar

deviasinya 0,82.

Fee memiliki nilai maksimum 28,29 yang berarti perusahaan klien yang

memberikan imbalan jasa paling besar dengan nilai logaritma natural 28,29 adalah

Astra International Tbk (ASII) pada tahun 2014.Nilai minimumnya 15,24 yang

185
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

berarti perusahaan klien yang memberikan imbalan paling kecil dengan nilai

logaritma natural 15,24adalah Lotte Chemical Titan Tbk, Titan Kimia Nusantara,

Fatra Polindo Nusa Industri Tbk (FPNI) pada tahun 2014. Rata-rata dari jumlah

sampel perusahaan klien memberikan imbalan dengan nilai logaritma natural

21,66dan standar deviasinya 2,09. Umur memiliki nilai maksimum 33 yang berarti

usia perusahaan yang tertua dan masih terdaftar adalah 33 tahun. Nilai minimum 1

yang berarti usia perusahaan yang paling muda sekaligus baru adalah 1 tahun. Rata-

rata dari umur perusahaan yang terdaftar adalah 19,49tahun dengan standar deviasi

7,57.

Tabel 2.
Hasil Kelayakan Model Hosmer and Lemeshow
Chi-square df Sig
3.481 8 0.901
Sumber: data diolah, 2016

Tabel 2 membuktikan bahwa nilai uji kelayakan model dari Hosmer dan

Lemeshow memiliki nilai chi-square sebesar 3,481 dan nilai signifikansinya adalah

0,901 yang lebih besar dari 0,05. Oleh karena itu hipotesis awal dapat diterima serta

model dapat memprediksi nilai observasinya.

Tabel 3.
Hasil Perbandingan Nilai -2 log Likelihood
Keterangan Nilai Log Likelihood (LL)
Nilai -2 LL awal (0) 277.525
Nilai -2 LL akhir (1) 153.233
Sumber: data diolah, 2016.

Tabel 3 membuktikan bahwa nilai -2 log likelihood awal sebesar 277,525

sedangkan nilai - 2 log likelihood akhir sebesar 153,233. Hal ini menunjukkan nilai

186
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

loglikelihood mengalami penurunan, maka model hipotesis yang dibuat adalah fit

dan cocok dengan data

Tabel 4.
Hasil Uji Koefisien Determinasi
-2 Log likelihood Cox & Snell R Square Nagelkerke R Square
153.233 0.461 0.616
Sumber: data diolah, 2016

Tabel 4 membuktikan bahwa nilai koefisien determinasi yang dijelaskan

melalui nagelkerke R square adalah 0,616. Nilai ini menunjukkan kombinasi seluruh

variabel bebas mampu menjelaskan varian dari variabel terikatnya sebesar 61,6%

lalu sisanya sebesar 38,4% dijelaskan oleh variabel lain diluar penelitian.

Tabel 5.
Hasil Uji Multikolinearitas
Konstanta Tenur Fee Rotasi Spesialisasi Umur
Konstanta 1.000 .000 -.001 .000 -1.000 .000
Tenur .000 1.000 -.188 .154 .000 -.070
Fee -.001 -.188 1.000 .431 .000 -.058
Rotasi .000 .154 .431 1.000 .000 .073
Spesialisasi -1.000 .000 .000 .000 1.000 .000
Umur .000 -.070 -.058 .073 .000 1.000
Sumber: data diolah 2016.

Tabel 5 membuktikan bahwa nilai koefisien korelasiantarvariabel bebas

memiliki nilai lebih kecil dari 0,8 sehingga model regresi ini baik karena tidak ada

gejala multikolinearitas yang serius antar variabel bebasnya.

Tabel 6.
Hasil Uji Matriks Klasifikasi
Observed Predicted
Kualitas Percentage Correct
0 1
0 87 21 80.6
Kualitas
1 18 75 80.6
Overall Percentage 80.6
Sumber: data diolah, 2016

187
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

Tabel 6 membuktikan bahwa kekuatan kemungkinan prediksi kualitas yang

diaudit oleh kap big 4 itu berkualitas baik adalah 80.6% dimana terdapat 75 dari 93

perusahaan yang diprediksi menggunakan jasa kap big 4. Kekuatan kemungkinan

prediksi kualitas yang tidak diaudit oleh kap big 4 itu berkualitas buruk adalah 80,6

% dimana terdapat 87 dari 108 perusahaan yang diprediksi tidak menggunakan jasa

kap big 4.

Tabel 7.
Hasil Uji Regresi Logistik
B S.E. Wald df Sig.
Tenur -1.106 .303 13.313 1 .000
Fee .755 .145 27.080 1 .000
Rotasi 1.554 1.054 2.174 1 .140
Spesialisasi 22.942 5693.556 .000 1 .997
Umur .007 .026 .065 1 .798
Konstanta -36.759 5693.557 .000 1 .995
Sumber: data diolah, 2016

Tabel 7 menjelaskan nilai koefisien model dari masing-masing variabel bebas

sehingga model regresi yang terbentuk adalah:


𝐾.𝐴
Ln ((1−𝐾.𝐴)) = -36,759 - 1,106Tenur + 0,755Fee + 1,554Rotasi +

22,942Spesialisasi + 0,007Umur + e

Hasil interpretasi dari model regresi tersebut dapat dijelaskan sebagai berikut:

Model memiliki nilai konstantanya sebesar -36,759. Apabila semua variabel

dianggap mempunyai nilai 0, maka kualitas audit yang dihasilkan cenderung

menurun sebesar 36,759.

Variabel X1Tenur memiliki nilai koefisien regresi sebesar -1,106. Hal ini

menandakan jika audit tenure cenderung berpengaruh negatif pada kualitas audit di

188
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

perusahaan manufaktur. Variabel X2Fee memiliki nilai koefisien regresi sebesar

0,755. Hal ini menandakan jika audit fee cenderung berpengaruh positif pada kualitas

audit di perusahaan manufaktur. Variabel X3Rotasi memiliki nilai koefisien regresi

sebesar 1,554. Hal ini menandakan jika rotasi auditor cenderung berpengaruh positif

pada kualitas audit di perusahaan manufaktur. Variabel X4 Spesialisasi memiliki nilai

koefisien regresi sebesar 22,942. Hal ini menandakan jika spesialisasi auditor

cenderung berpengaruh positif pada kualitas audit di perusahaan manufaktur.

Variabel X5Umur memiliki nilai koefisien regresi sebesar 0,007. Hal ini menandakan

jika umur publikasi cenderung berpengaruh positif pada kualitas audit.

PEMBAHASAN HASIL PENELITIAN

Berdasarkan tabel 7 disajikan pembahasan hasil pengujian hipotesis dalam penelitian

sebagai berikut: Hasil uji hipotesis pertama menunjukkan nilai signifikansi tenur

sebesar 0,000 dan kurang dari 0,05. Hal ini membuktikan hipotesis dapat diterima

yaitu audit tenure mempunyai pengaruh negatif pada kualitas audit. Adanya

pengaruh negatif juga diperkuat oleh bukti riset yang dilakukan Siregar et al., (2012)

dan Giri (2010). Alasannya karena jika melakukan perikatan yang lama dengan

auditor akan membuat ancaman bagi keobjektivitasnya dalam bekerja, sehingga tidak

muncul keberanian untuk mengungkap keadaan yang terjadi sebenarnya pada

perusahaan klien tersebut. Hasil uji hipotesis kedua menunjukkan nilai signifikansi

fee 0,000 dan kurang dari 0,05. Hal ini membuktikan hipotesis dapat diterima yaitu

audit fee mempunyai pengaruh positif pada kualitas audit. Adanya pengaruh positif

juga didukung oleh penelitian dari Maharani (2014) serta Kurniasih dan Rohman

189
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

(2014). Pemberian fee yang banyak kepada auditor memang terbukti membuat

auditor semakin rajin dan meningkatkan kinerjanya dalam bertugas, sehingga

nantinya dapat menyelesaikan pekerjaan sesuai prosedur dan menghasilkan kualitas

terbaik.

Hasil uji hipotesis ketiga menunjukkan nilai signifikansi sebesar 0,140 yang

lebih besar dari 0,05. Hal ini membuktikan hipotesis ditolak yaitu rotasi auditor tidak

mempunyai pengaruh pada kualitas audit. Pernyataan tidak berpengaruhnya rotasi ini

diperkuat oleh penelitian dari Hartadi (2012) dan Pamungkas (2014) yang

menyatakan pendapat serupa. Melakukan pergantian auditor memang dapat membuat

penyegaran namun auditor belum tentu mampu memahami keadaan perusahaan

kliennya sekarang dan memerlukan waktu untuk bisa mengerti keadaan lingkungan

baru tempatnya bekerja. Hasil uji hipotesis keempat menunjukkan nilai signifikansi

0,997 dan lebih dari 0,05. Hal ini membuktikan hipotesis ditolak yaitu spesialisasi

auditor tidak memberikan pengaruh pada kualitas audit. Pernyataan tidak

berpengaruhnya spesialisasi didukung oleh penelitiannya Mahetmi (2012), Novianti

(2012), dan Rusmin (2010). Alasannya adalah jika perusahaan klien tidak

menggunakan auditor spesialis dan lebih memilih auditor yang berasal kap besar atau

big 4 maka akan memiliki kualitas yang sama bagusnya, sehingga penggunaan

spesialis auditor tidak terlalu berdampak pasti.

Hasil uji hipotesis kelima menunjukkan nilai signifikansi sebesar 0,798 yang

lebih besar dari 0,05. Hal ini membuktikan hipotesis ditolak yaitu umur publikasi

tidak berpengaruh pada kualitas audit. Adanya penolakan tersebut didukung oleh

190
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

penelitian Paramita dan Latrini (2015) yang mendapatkan hasil serupa. Umur tidak

memberikan jaminan bahwa perusahaan yang berusia tua memiliki kualitas yang

lebih bagus dari perusahaan muda, sehingga umur tidak tepat untuk dijadikan

patokan dalam menghasilkan kualitas yang bagus.

SIMPULAN DAN SARAN

Pembahasan hasil penelitian, maka simpulan yang dapat dibuat adalah audit tenure

dapat memberi pengaruh negatif, audit fee dapat memberi pengaruh positif, umur

publikasi, rotasi auditor, dan spesialisasi auditor tidak memberikan pengaruh pada

kualitas audit perusahaan manufaktur di Bursa Efek Indonesia periode 2013-2015.

Saran yang dapat diberikan kepada peneliti berikutnya sesuai simpulan tersebut

adalahmenambahkan jumlah periodepengamatandalam riset yaitu selama lebih dari 3

tahun, menambahkan beberapa variabel bebas untuk menyempurnakan penelitian ini,

dan memperluas sektor industri yang dijadikan populasi agar mendapat hasil yang

pasti.

REFERENSI

Abdelrahman, A. N., Khaled, Z., Elbayoumi, A. F. 2014. Measuring audit firm


intellectual capital as a determinant of audit quality: suggested model.Journal
of Modern Accounting and Auditing. Vol.10 pp.59-79

Al-Thuneibat. 2011. Do audit tenure and firm size contribute to audit quality?
empirical evidence from Jordan. Journal of Managerial Auditing. Vol.26
pp.317-334

Baotham, S. 2007. The impact of professional knowledge and personal ethic on audit
quality. Journal of Internasional Academy Business and Economi.

191
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

Bawono, S. 2010. Faktor-faktor dalam diri auditor dan kualitas audit: studi pada
KAP Big 4 di Indonesia. Universitas Negeri Jenderal Soedirman,
Purwokerto.

Bestivano. 2013. Pengaruh ukuran perusahaan, umur perusahaan, profitabilitas, dan


leverage terhadap perataan laba pada perusahaan perbankan di BEI. Fakultas
Ekonomi Universitas Negeri Padang. Sumatera Barat.

Chen, C. Y., and Lin, Y. C. 2004. Audit partner tenure, audit firm tenure, and
discrestional accruals: does long auditor tenure impair earnings quality?.
Hongkong University of Science and Technology.

DeAngelo. 1981. Auditor independence, lowballing, and disclosure regulation.


Journal of Accounting and Economic. Vol.3 pp.113-127

Edosa, Aronmwan, J., Ashafoke, T. O., Mgbame, C. O. 2013. Audit firm reputation
and audit quality. European Journal of Business and Management. Vol.5
No.7

Enofe, D. O., Nbgame, D. C., Ediae, C.O. 2013. Audit quality and auditors
independence in Nigeria: an empirical evaluation research. Journal of
Finance and Accounting. Vol.4

Fernando, G. D. 2010. Audit quality attributes, client Size and cost of equity
capital. Review of Accounting and Finance. Vol.9 No.4 pp. 363-381

Fitriany. 2011. Pengaruh tenure audit dan auditor spesialis terhadap informasi
asimetri. Universitas Indonesia. Skripsi.

Ghozali, I. 2012. Aplikasi Analisis Multivariate dengan Program IBM SPSS 20.
Badan Penerbit Universitas Diponegoro. Semarang.

Giri, E. F. 2010. Pengaruh tenur KAP dan reputasi KAP terhadap kualitas audit:
kasus rotasi wajib auditor di Indonesia. Simposium Nasional Akuntans XIII.
Purwokerto.

Gul, F. S., dan Jaggi, B. 2008. Earnings quality: some evidence on the role of
auditor tenure and auditors industry expertise. Journal of Accounting and
Economic. Vol.6 No.1

Hamid, A. 2013. Pengaruh tenur KAP dan ukuran KAP terhadap kualitas audit
perusahaan manufaktur di BEI. Jurnal akuntansi. Vol.1 No.1 Universitas
Negeri Padang. Sumatera Barat.

192
I Gusti Ngurah Indra Pramaswaradana dan Ida Bagus Putra Astika. Pengaruh Audit….

Hanjani, A. 2012. Pengaruh etika auditor, pengalaman auditor, fee audit, dan
motivasi auditor terhadap kualitas audit. Universitas Diponegoro. Semarang.
Skirpsi.

Hartadi, B. 2012. Pengaruh fee audit, rotasi KAP, dan reputasi auditor terhadap
kualitas audit di BEI. Jurnal Ekonomi dan Keuangan. Vol.16 No.1 pp.84-103

Herusetya, A. 2009. Pengaruh ukuran auditor dan spesialisasi auditor terhadap


kualitas laba. Jurnal Akuntansi dan Keuangan Indonesia. Vol.6 No.1
Universitas Pelita Harapan. Jakarta

Hidayat, dan Elizabet. 2007. Faktor-faktor yang mempengaruhi kualitas pelaporan


keuangan pada perusahaan manufaktur yang Go Publik di Indonesia.
Universitas Airlangga.

Jensen, M. C., dan Meckling, W.H. 1976. Theory of the firm: managerial behavior,
agency cost, and ownership structure. Journal of Financial Economics. Vol.3
No.4 pp.305-360

Kurniasih, M., dan Rohman, A. 2014. Pengaruh audit fee, audit tenure, dan
rotasi audit terhadap kualitas audit. Diponegoro Journal of Accounting.
Vol.3 No.3

Maharani, A. P. 2014. Pengaruh audit tenure, audit fee, rotasi audit, spesialisasi audit
terhadap kualitas audit. Universitas Muhamadyah. Surakarta.

Mahetmi, I. S. 2012. Pengaruh tenur kap, reputasi kap, spesialisasi industri kap dan
rotasi kap, terhadap kualitas audit. Universitas Muhamaddyah. Yogyakarta.

Meneri Keuangan. 2008. Peraturan Menteri Keuangan Nomor 17/PMK.01/2008


Tentang Jasa Akuntan Publik. Jakarta.

Myers. 2013. Exploring the term of the auditor client relationship and the quality of
earnings. The Accounting Review. Vol.78 No.3 pp.779-799

Novianti. 2012. Pengaruh tenur kap, tenur partner, asuditor spesialis industri
terhadap kualitas audit. Simposium Nasional Akuntansi. Balikpapan.

Pamungkas, Y. C. 2014. Pengaruh fee audit, rotasi kap, dan reputasi auditor terhadap
kualitas audit. Jurnal Riset Manajemen dan Akuntansi. Vol.2 No.2 pp.59-67

193
ISSN: 2302-8556
E-Jurnal Akuntansi Universitas Udayana
Vol.19.1. April (2017): 168-194

Panjaitan. 2014. Pengaruh tenure, ukuran kap, dan spesialisasi auditor terhadap
kualitas audit. Diponegoro Journal of Accounting. Vol.3 No.3

Paramita, A., dan Latrini, Y. 2015. Pengaruh ukuran perusahaan, umur publikasi,
masa perikatan audit, dan pergantian manajemen pada kualitas audit. E-
Jurnal Akuntansi. Universitas Udayana. Vol.13 No.1 pp.142-156

Perdana, M. A. 2014. Pengaruh rotasi kantor akuntan publik dan rotasi partner
auditor terhadap kualitas audit. Universitas Diponegoro. Skripsi.

Permana, K. X. 2012. Pengaruh masa perikatan audit dan ukuran KAP terhadap
kualitas audit. Universitas Diponegoro. Skripsi.

Romanus, Maher, Flaming. 2008. Auditor industryspecialization, auditor changes,


and accounting restatement. Accounting Horizon. Vol.22 No.22 pp.389-
413

Rusmin. 2010. Auditor quality and earnings management Singapore eviden.


Journal of Managerial Auditing. Vol.25 No.7 pp.618-638

Sinaga, D. 2012. Analisis pengaruh audit tenure, ukuran kap, dan ukuran
perusahaan klien terhadap kualitas audit. Universitas Diponegoro. Skripsi.

Siregar, Veronica, S., Amarulah, F., Wibowo, A. 2012. Audit tenure, audit rotation,
and audit quality. Asian Journal of Business and Accounting. Vol.5 pp.55-74

Tan, P. 2000. Business excellent in Enterpreneurship though motivation audit.


Managerial Auditing Journal. Vol.15 No.7 pp.338-347

Velte, P., dan Stiglbauer, M. 2012. Impact of auditor and audit firm rotation on
accounting and audit quality. International Conference Improfing Financial
Institusion.

194

Anda mungkin juga menyukai