The European Accounting Review 1998, 7:4, 623–635

International transfer pricing and

income shifting: evidence from
the UK
P. B. Oyelere
Lincoln University, New Zealand
C. R. Emmanuel
University of Glasgow

ABSTRACT
The potential use of international transfer pricing (ITP) as an income-shifting
mechanism by multinational enterprises (MNEs) has long been recognized.
However, there is relatively little evidence to substantiate or discount this claim in
relation to UK-based foreign-controlled enterprises (FCEs). This paper examines
the possible use of ITP as an income-shifting mechanism by FCEs operating in the
UK. The methodological approach involves the comparison of the proŽ tability
(performance) and dividend (post-performance) distributions of a sample of FCEs
with those of UK-controlled enterprises (UKCEs) over a two-year period. The two
samples are matched on the basis of their total assets (capability). Results reveal
signiŽ cant differences in the proŽ tability and dividend distributions of the two
groups. FCEs underperform UKCEs, but their level of dividend distribution
outstrips those of UKCEs. Based on this sample of seventy-two companies, a Ž rm
is more likely to be an FCE, rather than a UKCE, if it reports a combination of
lower performance and higher post-performance distribution. Evidence of sig-
niŽ cant income shifting by FCEs is conŽ rmed and the claim that ITP is the key
mechanism for such shifts cannot be dismissed.

INTRODUCTION
This paper investigates the link between reported proŽ tability and dividend
distributions of UK-based foreign-controlled enterprises (FCEs) and UK-
controlled enterprises (UKCEs) to determine whether international transfer
pricing (ITP) is used for income-shifting purposes. ITP is concerned with the
monetary value attaching to movements of goods and services between parts
of the same enterprise which cross national boundaries. Its potential use as an
income-shifting mechanism by multinational enterprises (MNEs) has long

Address for correspondence

Peter B. Oyelere, Accountancy Group, Commerce Division, PO Box 84, Lincoln
University, New Zealand. E-mail: [email protected]

© 1998 European Accounting Association 0963–8180

624 The European Accounting Review

been recognized (Picciotto, 1992). Empirical evidence is, however, lacking

with respect to FCEs’ transfer pricing practices in the UK.
Positive correlation should exist between Ž rms’ capability, proŽ tability
and dividend distribution (Wheeler, 1990). Proceeding on this logical as-
sumption, a sample of FCEs is matched with a sample of UKCEs on the basis
of capability, and consequently, a comparison of the reported proŽ tability
(performance) and dividend (post-performance) distributions of the two
samples is undertaken. The results reveal signiŽ cant divergence in the
performance and post-performance distributions of the two samples. We Ž nd
an unusual relationship between the proŽ tability and distributions of FCEs
which suggests that transfer pricing is used to shift income from the UK.
Logit regression estimates enable the headquarter or control location of Ž rms
to be predicted from a number of performance and post-performance explan-
atory variables. A Ž rm is more likely to be an FCE when lower proŽ ts and
higher dividend distributions are reported.
The rest of this paper is organized as follows. The next section reviews the
literature on ITP and income shifting. The chosen research method is then
explained and a subsequent section discusses the empirical results. Finally,
the conclusions and implications for future research are reported.

INTERNATIONAL TRANSFER PRICING AND INCOME

SHIFTING
One of the potential roles of ITP is to locate group proŽ ts artiŽ cially in
countries where a tax advantage may be obtained. By manipulation of ITP,
global tax liability can be minimized. Several well-known texts, such as
Radebaugh and Gray (1997), Rugman and Hodgetts (1995) and Shapiro
(1992) illustrate how this is achieved. Yet the issue is not merely one of
academic or theoretical interest. Newspaper headlines like ‘IRS investigates
foreign companies for tax cheating’ (New York Times, 18 February 1990),
‘Inland Revenue probes tax avoidance at Sony’ (Sunday Times, 22 March
1992), ‘Apple Computer Ž ghts claim by IRS’ (New York Times, 3 April 1993)
and ‘Transfer pricing comes out of the shadows’ (Financial Times, 23
November 1995) testify to the practical signiŽ cance. Research on the
income-shifting role of ITP is, however, relatively scarce and mainly
conŽ ned to the USA.
Grubert et al. (1993), Harris et al. (1991) and Scholes et al. (1992) report
evidence of income shifting by US-based Ž rms in response to tax incentives
and differentials. When tax rates, bases and rules differ across countries, it is
generally not a matter of indifference where income is reported. Hence
MNEs in high-tax countries tend to report higher US income and those in
low-tax foreign countries tend to report lower US income. Harris (1993),
Jacob (1996) and Klassen et al. (1993) Ž nd that US-based MNEs shift
income between geographic locations in response to changes in tax rates and
 International transfer pricing and income shifting 625

rules. The volume of intra-Ž rm international transfers appears to suggest that

transfer prices are the mechanism for income shifting. However, direct
evidence of ITP playing this role is difŽ cult to obtain due to the sensitivity of
the issue for both MNEs and Ž scal agencies.
An indirect method to detect income shifting is applied by Crain and Stitts
(1994), Kim and Lyn (1990) and Wheeler (1988, 1990). Taking matched
samples of FCEs and US-controlled enterprises, performance as measured by
gross proŽ t margin or rate of return on assets or some other proŽ tability
index is compared. Each of these studies reported the performance of US-
controlled enterprises to be signiŽ cantly higher than those of FCEs. Wheeler
(1988) extended the analysis to investigate post-performance distributions of
FCEs and US-controlled enterprises. For the period 1972–83, he found
that:
1 the rate of return on assets for US-controlled enterprises was six times
larger than that reported by FCEs;
2 distributions to shareholders by FCEs increased sixfold whilst the FCEs’
return on assets declined over this time period.
This study obtained data from the Statement of Income Ž led by FCEs with
the Internal Revenue Service. It is noteworthy that the number of statements
FCEs Ž led increased substantially during this time span. Wheeler concluded
that improper transfer-pricing policies appeared to be the only plausible
explanation for the discrepancies discovered.
However, an alternative explanation for differences in performance be-
tween FCEs and US-controlled enterprises relates to operational and strategic
factors. Managerial skill and experience, start-up costs, age of investment,
nature of the product and the production process are some of the reasons why
performance differs (Gideon, 1990). Additionally, Buckley and Hughes
(1997) argue that differing cost principles, as illustrated by the use of target
costing in Japanese enterprises, justify the application of transfer prices
which are perceived as ‘fair’ for that national business culture. The sig-
niŽ cance of these arguments is likely to be greater when individual FCEs and
UKCEs are compared. When samples of FCEs and UKCEs are the focus of
analysis, the effect of operational, strategic and cultural factors is likely to be
mitigated.
Occasional anecdotal evidence (Sunday Times, 22 March 1992) indicates
that differences in performance exist between FCEs and UKCEs. British
MNEs paid more than Ž ve times as much tax per pound of turnover than their
Japanese counterparts. In 1988–89 and 1990–91, the Inland Revenue col-
lected less than £40 million in proŽ ts tax from the ten largest Japanese Ž rms
in the UK which represented only 0.6% of their combined UK turnover. This
is in stark contrast to the £14 billion proŽ ts tax which the largest ten British
companies paid during the same period. Whilst culture and other operational
or strategic factors may in uence performance, the potential signiŽ cance of
626 The European Accounting Review

ITP cannot be ignored. Using UK data, this study investigates the use of ITP
as an income-shifting mechanism.

RESEARCH METHOD
Sample selection and data collection
The aim of this study is to determine whether the proŽ tability (performance)
and dividend (post-performance) distributions of FCEs operating in the UK
reveal evidence of the use of ITP for income shifting. The approach
employed involves a comparison of the proŽ tability and dividend distribu-
tions of a sample of these FCEs with those of their UK-controlled counter-
parts. The underlying assumption is that capability (as measured by total
assets) is related to performance (as measured by returns) which in turn relate
to post-performance events (as measured by dividend payouts). 1 The ap-
proach is applied to a sample of thirty-six FCEs operating in the UK which
are matched with thirty-six UKCEs on the basis of total asset value. Within
the constraint of total asset values for each sample being equal in 1992 and
1993, an effort to create a UKCE sample which re ected the industry sector
composition of the FCEs is made. By controlling the capability of the two
samples over the period of analysis, similarities and differences in perform-
ance and post-performance are observed.
One hundred and forty FCEs were randomly selected from the Times 1000
(1994), Kompass or Dun & Bradstreet’s Who Owns Whom (1993). A letter
was mailed to each of them requesting copies of their 1992 and 1993 annual
reports and accounts. Seventy-nine responses were received. Thirty-three of
the respondents provided the reports of their parent companies which
presented Ž nancial data in their home currency and were therefore not usable.
Seven respondents contended that, being private companies, they are not
required by law, and were therefore unwilling, to agree to our request. Three
respondents gave other reasons for not providing their reports. The remaining
thirty-six respondents provided usable 1992 and 1993 annual reports for the
purpose of our study. Usable means that the reports are in English, denom-
inated in pounds sterling and identify the Ž rm’s UK operations.
The next step was to match the thirty-six FCEs with UKCEs of similar
capability. 2 Total assets was used as the measure of capability. A total of
seventy-two Ž rms (thirty-six FCEs and thirty-six UKCEs) thus make up our
sample. An industry sector breakdown of sampled Ž rms is provided in Table
1. Industrial match is inexact with pharmaceuticals unrepresented in the UK-
owned sample and engineering and construction possibly being overrepre-
sented. Future studies may need to alleviate this, but it should be recognized
that all of the companies sampled are sufŽ ciently large to exhibit some
degree of diversiŽ cation. Hence, the industry classiŽ cation based on the main
activity is itself inexact. The apparent disparity in the UK sample was
 International transfer pricing and income shifting 627

Table 1 FCEs and UKCEs by industry

FCEs UKCEs

Engineering and construction 1 6

Services including banking and Ž nancial services 12 11
Electronics and electrical 2 5
Computer and information technology 5 3
Retail and departmental stores 2 1
Photographics 1 1
Food, beverages, etc. 3 5
Fuel and energy 1 3
Manufacturing 5 1
Pharmaceuticals 4 –
36 36

necessitated by the need to equate the asset values in aggregate with those of
the foreign-controlled sample of enterprises for both 1992 and 1993.
For each company, summary data were extracted on their:
(i) Capability as measured by total assets. For the UK sample, use was
made of the segment report relating to UK assets.
(ii) Performance as measured by proŽ t before taxation and turnover.
Again, use was made of the UK data in the segment report.
(iii) Post-performance events as measured by the absolute amount of
dividend paid, and dividend per share.
A summary of the data obtained is presented in Table 2. Preliminary
overview of this data reveals a number of points. First, the two samples are
similar in terms of their collective capability in that they hold similar
amounts of total assets in the same location, namely the UK. No difference
is revealed in the measures of capability for the two samples. This indicates
that the matching process was successful and our two samples are therefore
considered to possess identical capabilities.
Second, from the performance data it is clear that differences, at least in
nominal terms, exist between the two groups. This difference holds good for
all the ratios and in both time periods. The Ž gures in Table 2 indicate that the
performance of FCEs is lower than that of UKCEs. For example, the
turnover generated by FCEs, with a mean of £738 million in 1993, is only
66% of the £1,116 million average for UKCEs. One possible explanation is
that FCEs collectively are experiencing difŽ culty in gaining market share in
the UK. Another possibility is that sales revenue is being underpriced for
intra-group trade. Similar differences exist in the net income of the two
groups. UKCEs consistently reported greater income than their foreign
counterparts. Furthermore, when income is expressed over total assets (return
on total assets), the mean return for FCEs is 2 20% and 2 226% as compared
628 The European Accounting Review

Table 2 Preliminary univariate data comparison

Variables Location 1993 1992

of
control Cases Mean Cases Mean

1. Capability
Total assets Foreign 36 £1,013m 36 £1,025m
UK 36 £1,010m 36 £1,010m
2. Performance
Total turnover Foreign 34 £738m 34 £698m
UK 36 £1,116m 36 £1,102m
Net income Foreign 36 2 £90m 36 2 £165m
UK 36 £84m 36 £55m
Return on total assets Foreign 36 2 20% 36 2 226%
UK 36 8% 36 7%
Return on total turnover Foreign 34 2 20% 34 2 79%
UK 36 7% 36 5%
3. Post-performance
Amount of dividend payout Foreign 28 £847m 27 £1,796m
UK 34 £28m 34 £24m
Dividend per share Foreign 31 £0.3924 29 £0.8577
UK 35 £0.0945 35 £0.0935
Dividend per £ of total turnover Foreign 28 £1.7878 27 £5.7603
UK 34 £0.0232 34 £0.0217
Dividend per £ of total assets Foreign 28 £2.2850 27 £8.1472
UK 34 £0.0322 34 £0.0264

to the 8% and 7% of UKCEs in 1993 and 1992, respectively. Such lower

returns may be achieved through the overpricing of UK-bound goods and
services. There is an expectation therefore that the ‘poor’ performance of
FCEs is re ected in their post-performance distributions.
Third, post-performance data reveal an opposite trend to performance.
Despite the poor performance reported by FCEs, their mean dividend payout
was signiŽ cantly higher than UKCEs’. For example, FCEs’ average payout
of £1,796 million in 1992 compares to UKCEs’ £24 million. This practice
suggests that cash  ow is not adversely affected by operational losses; that is,
dividend does not appear to be a function of income. UKCEs exhibited a
more traditional pattern with average dividend per share holding steady at 9
pence per share for both years, whilst those of FCEs  uctuated wildly, with
a high of 86 pence in 1992 and 39 pence in 1993.
In summary, clear patterns may be detected from the above data. On the
one hand, FCEs report lower performance in comparison to UKCEs, while,
on the other hand, their post-performance activities suggest the exact
opposite as their dividend payouts clearly outstrip those of their domestic
counterparts. This performance and post-performance divergence requires
further examination.
 International transfer pricing and income shifting 629

Statistical analysis
To investigate differences between FCEs and UKCEs performance and post-
performance, a regression methodology is utilized. In this instance, the
dependent variable is necessarily classiŽ ed as a binary choice variable (that
is, a company is allocated a value of 1 if it is an FCE and 0 if it is a UKCE).
Given the dichotomous nature of the dependent variable, a logit model is
employed in this study (Dietrich and Sorensen, 1984). Logit analysis, while
accounting for the stochastic element in the outcome attributable to the
disturbance distributions, relates the probability of an event to some measur-
able Ž rm performance factors. The estimation allows a comparison of the
relative importance of the explanatory variables.
The model we estimated is algebraically stated as:

Yi 5 a 1 O Xijb 1 m i (1)

where, for the ith Ž rm,

Y 5 the dependent variable which is 1 if the Ž rm is foreign-controlled and
0 if the Ž rm is UK-controlled;
a 5 the intercept of the equation;
Xj 5 the observation on the independent variable j;
b 5 the coefŽ cient of the independent variable to be estimated from the
data;
m 5 stochastic disturbance term.
In this paper, P(Yi) is the probability of being foreign-owned or UK-
controlled; Xi’s are the capability, performance and post-performance meas-
ures of sampled Ž rms; a and b are the parameters to be estimated.
Stated in its full form, our logit estimation equation is:
Yi 5 a 1 b 1(Capability measure) i 1 b 2(Performance measure)
1 b 3(Post-performance measure) 1 m i (2)
where, for Ž rm i,
Y5 location of Ž rm’s headquarters; 1 for FCE and 0 for UKCE;
m 5 error term.
From the overview of data presented in Table 2, the a priori expectation is
that the coefŽ cient estimate on capability ( b 1) should be insigniŽ cantly
different from zero; that the predicted sign on b 2 should be negative, while b 3
should be positively signed.

FINDINGS
Equation (2) is estimated using alternative speciŽ cations for the performance
and post-performance variables (as detailed in Table 2). An overview of the
630 The European Accounting Review

coefŽ cient estimate and associated statistics is provided in Tables 3a, 3b and
4a, 4b. For the results in Tables 3a and 3b, the model is speciŽ ed as follows:
the capability measure is total assets; the performance measure is net income;
while the post-performance measure is dividend per share. Tables 3a and 3b
differ in that the former refers to 1992 data while the latter is based on 1993
data.
Similarly, the speciŽ cation in Tables 4a and 4b is as follows: total assets,
is again used as the capability measure; the performance measure is return on
total assets; and amount of dividend payout is employed as the post-
performance measure. As before, Table 4a is based on 1992 data, while Table
4b relates to 1993 information. Alternative speciŽ cations of performance and
post-performance using turnover variables are avoided to limit the possible
in uence of ITP. Across the respective speciŽ cations, a broadly similar
picture emerges. First, as expected, the coefŽ cient estimate on the capability
measure (total assets) is insigniŽ cantly different from zero. This result
indicates that the matching process in this respect was successful.
Second, in line with preliminary data results detailed in Table 2 and a
priori prediction, the coefŽ cient signs of performance variables are negative
as predicted. It should be noted that the estimate in Table 3a is signiŽ cant at
the 5% level, while those in Tables 3b, 4a and 4b are signiŽ cant at the 10%
level. These results support our contention that despite matched capability,
FCEs signiŽ cantly underperformed UKCEs in both years.
Third, in both 1992 and 1993, the coefŽ cients of post-performance
variables returned positive signs as predicted. The estimate of dividend per

Table 3a Results of logit estimate: 1992 data

Variablea CoefŽ cient S.E. Sig.

CAPABLE2 0.3283 0.8258 .3077

PERFORM2 2 1.4545 0.7121 .0411 **
POST_PERF2 2.1263 2.0698 .1043
CONSTANT 6.3167 4.0706 .1207
Number of observations 64b
Log likelihood 85.11%
Goodness of Ž t 68.83%
Correctly predicted: FCEs 96.67%
UKCEs 44.11%
Overall 75.51%
Notes:
** Denotes signiŽ cant at 5% level.
a
The dependent variable is Y, which is 1 if the Ž rm is foreign-controlled and 0 if the Ž rm is
UK-controlled. The independent variables are: CAPABLE2 5 Capability measure (Total assets)
in 1992; PERFORM2 5 Performance measure (Net income) in 1992; POST_PERF2 5 Post-
performance measure (Dividend per share) in 1992.
b
Eight cases were rejected because of missing data, leaving 64 valid cases.
 International transfer pricing and income shifting 631

Table 3b Results of logit estimate: 1993 data

Variablea CoefŽ cient S.E. Sig.

CAPABLE3 0.7919 0.7269 .2760

PERFORM3 2 1.1042 0.6307 .0800*
POST_PERF3 2.8780 1.6033 .0726*
CONSTANT 5.3629 3.4261 .1175
Number of observations 66b
Log likelihood 66.58%
Goodness of Ž t 61.43%
Correctly predicted: FCEs 93.75%
UKCEs 48.00%
Overall 73.68%
Notes:
* Denotes signiŽ cant at 10% level.
a
The dependent variable is Y, which is 1 if the Ž rm is foreign-controlled and 0 if the Ž rm is
UK-controlled. The independent variables are: CAPABLE3 5 Capability measure (Total assets)
in 1993; PERFORM3 5 Performance measure (Net income) in 1993; POST_PERF3 5 Post-
performance measure (Dividend per share) in 1993.
b
Six cases were rejected because of missing data, leaving 66 valid cases.

share in 1993 (Table 3b) is signiŽ cant at the 10% level. These results indicate
that in spite of lower levels of performance, FCEs made higher dividend
distributions in 1992 and 1993.
To summarize the three points above, it is obvious that despite having
similar capability, FCEs underperformed their UK-controlled counterparts;

Table 4a Results of logit estimate: 1992 data

Variablea CoefŽ cient S.E. Sig.

CAPABLE2 2 0.0008 0.0002 .7413

PERFORM2 2 1.0252 0.0453 .0777*
POST_PERF2 0.5857 0.4328 .1760
CONSTANT 2 4.5788 2.9411 .1195
Number of observations 61b
Log likelihood 72.50%
Goodness of Ž t 60.86%
Correctly predicted: FCEs 97.06%
UKCEs 62.96%
Overall 81.97%
Notes:
* Denotes signiŽ cant at 10% level.
a
The dependent variable is Y, which is 1 if the Ž rm is foreign-controlled and 0 if the Ž rm is
UK-controlled. The independent variables are: CAPABLE2 5 Capability measure (Total assets)
in 1992; PERFORM2 5 Performance measure (Return on total assets) in 1992; POST_
PERF2 5 Post-performance measure (Dividend amount) in 1992.
b
Eleven cases were rejected because of missing data, leaving 61 valid cases.
632 The European Accounting Review

Table 4b Results of logit estimate: 1993 data

Variablea CoefŽ cient S.E. Sig.

CAPABLE3 2 0.0001 0.0002 .6606

PERFORM3 2 1.0147 0.0247 .0617 *
POST_PERF3 0.6569 0.4976 .1868
CONSTANT 2 5.1050 3.3818 .1312
Number of observations 62b
Log likelihood 74.88%
Goodness of Ž t 62.01%
Correctly predicted: FCEs 97.06%
UKCEs 60.71%
Overall 80.65%
Notes:
* Denotes signiŽ cant at 10% level.
a
The dependent variable is Y, which is 1 if the Ž rm is foreign-controlled and 0 if the Ž rm is
UK-controlled. The independent variables are: CAPABLE3 5 Capability measure (Total assets)
in 1993; PERFORM3 5 Performance measure (Return on total assets) in 1993; POST_
PERF3 5 Post-performance measure (Dividend amount) in 1993.
b
Ten cases were rejected because of missing data, leaving 62 valid cases.

their post-performance distributions, however, suggest the opposite as greater

dividend payouts were made by FCEs. The predictive implication of these
Ž ndings is that, all things being equal, the probability that a Ž rm is foreign-
controlled increases as it reports a combination of lower performance and
higher post-performance distributions.

CONCLUSIONS
This study investigates the possible use of ITP as an income-shifting
mechanism by FCEs operating in the UK. Based on the assumption that
capability (assets) and performance (proŽ tability) should be positively corre-
lated, with similar implications for post-performance events (dividend pay-
outs), a comparison of the Ž nancial data of a sample of FCEs and UKCEs is
undertaken. Our Ž ndings reveal differences between the performance and
post-performance activities of the two samples. From the results, it may be
possible to predict the control location (that is, either foreign- or UK-
controlled) of sampled Ž rms on the basis of their level of proŽ tability and
dividend payouts. The performance and post-performance activities of FCEs
reveal an unusual relationship. Explanation of the differences is open to
conjecture. At this stage, it is difŽ cult to reject ITP manipulation as a
plausible explanation for lower reported income, especially as these low
performances do not appear to hinder superior or equal post-performance
distributions by FCEs in comparison with UKCEs. There appears to be prima
facie evidence of income shifting through ITP. This merits further
investigation.
 International transfer pricing and income shifting 633

However, two constraints should be recognized for future research in this

area. First, the need for any future study to ensure a more rigorous match
between the enterprises sampled. This may recognize industry sectors,
experience in the location, accounting disclosure compliance and time span
of the study. In this last respect, a period of investigation which extends
beyond two years seems merited. By these means, the methodological design
of the current study may be improved.
The second constraint relates to the Ž scal authority where the investigation
takes place. The UK has long adopted the ‘arm’s-length’ principle for ITP
(Taxes Management Act, 1970), and possesses the power to substitute any
transfer price which is believed to distort taxable income in the UK.
However, the rules do not provide speciŽ c recommendations as in the USA
and full texts of court cases are not available. Past tax audit investigations in
this area are also veiled in conŽ dentiality. The presumption is that Inland
Revenue inspectors through their regular discussions with MNEs will effec-
tively convey what is acceptable as a transfer pricing policy. This has certain
real merits, not least being the development of a co-operative relationship
which may ensure that relevant information is made available to the Ž scal
authority in a timely fashion. However, this development of co-operation
requires conŽ dentiality to be upheld and, consequently, a near-absolute
embargo on information reaching the public domain.3 Seeking evidence of
ITP manipulation is a legitimate issue for academic enquiry. It is unfortunate
that it must be undertaken in an indirect manner.

ACKNOWLEDGEMENTS
The authors gratefully acknowledge the comments of participants at the
Tenth Irish Accounting and Finance Conference, at Dublin City University
on 8–9 May 1997, on an earlier draft of this paper. We also thank Professor
Donal McKillop for his help and encouragement.

NOTES
1 For an elaborate discussion of this assumption, see Wheeler (1988, 1990).
2 The UKCEs were selected from a list of UK Ž rms whose annual reports and
accounts for 1992 and 1993 are archived in Glasgow University. Each UKCE was
closely matched with a respondent FCE on the basis of total assets.
3 The recent Inland Revenue consultative document on ‘Modernisation of the
Transfer Pricing Legislation’ (1997) may have an in uence but there is little
likelihood that the introduction of self-assessment will result in publicly available
data.

REFERENCES
Buckley, P. J. and Hughes, J. F. (1997) ‘Japanese transfer pricing policy: a note’,
Applied Economic Letters, 4: 13–17.
634 The European Accounting Review

Crain, T. L. and Stitts, R. H. (1994) ‘A comparison of gross proŽ t margins between

foreign-controlled domestic corporations and comparable US-controlled domestic
corporations’, Journal of International Accounting Auditing & Taxation, 3(1):
85–101.
Dietrich, J. K. and Sorensen, E. (1984) ‘An application of logit analysis to prediction
of merger targets’ Journal of Business Research, 12: 393–402.
Emmanuel, C. R. (1977) ‘Transfer pricing: a diagnosis and possible solution to
dysfunctional decision-making in the divisionalised company’, Management Inter-
national Review, 17(4): 45–9.
Emmanuel, C. R. and Mehafdi, M. (1994) Transfer Pricing. London: Academic
Press.
Ernst & Young (1993) Rethinking Intragroup Services: The Con ict Between Theory
and Practice in Transfer Pricing. London: E&Y Thought Leadership series.
Financial Times (1995) ‘Transfer pricing comes out of the shadows’, 23 November,
p. 11.
Gideon, K. W. (1990) ‘Statement before the Ways and Means Oversight Sub-
committee Hearing on foreign-owned businesses in the US’, in Daily Tax Report,
13 June, pp. L25–9.
Grubert, H. and Mutti, J. (1991) ‘Taxes, tariffs and transfer pricing in multinational
corporate decision making’, Review of Economics and Statistics, 73: 285–93.
Grubert, H., Goodspeed, T. and Sivenson, D. (1993) ‘Explaining the low taxable
income of foreign-controlled companies in the United States’, in Giovanni, A.,
Hubbard, G. and Slemrod, J. (eds) Studies in International Taxation, National
Bureau of Economic Research Project Report, University of Chicago Press.
Harris, D. G. (1993) ‘The impact of US tax law revision on multinational corpora-
tions’ capital location and income shifting decisions’, Journal of Accounting
Research, 31 (Supplement): 111–39.
Harris, D., Morck, R., Slemrod, J. and Yeung, B. (1991) ‘Income shifting in US
multinational corporations’, Working Paper, NBER no. 3924.
Jacob, J. (1996) ‘Taxes and transfer pricing: income shifting and the volume of
intraŽ rm transfers’, Journal of Accounting Research, 34(2) (Fall): 301–15.
Kim, W. S. and Lyn, E. O. (1990) ‘FDI theories and the performance of foreign
multinationals operating in the US’, Journal of International Business Studies,
First quarter: 41–54.
Klassen, K., Lang, M. and Wolfson, M. (1993) ‘Geographic income shifting by
multinational corporations in response to tax rate changes’, Journal of Accounting
Research, 31 (Supplement): 141–73.
New York Times (1990) ‘IRS investigates foreign companies for tax cheating’, 18
February, pp. 1 and 30.
New York Times (1993) ‘Apple computer Ž ghts claim by IRS’, 3 April.
OECD (1993) Tax Aspects of Transfer Pricing within Multi-national Enterprises:The
United States Proposed Regulations, Committee on Fiscal Affairs on the Proposed
Regulations under Section 482 IRC. Paris: OECD.
Picciotto, S. (1992) ‘International taxation and intra-Ž rm pricing in trans-national
corporate groups’, Accounting, Organisations and Society, 17(8): 759–92.
Radebaugh, L. and Gray, S. (1997) International Accounting and Multinational
Enterprises, 4th edn. Chichester: John Wiley.
Rugman, A. and Hodgetts, R. (1995) International Business: A Strategic Manage-
ment Approach. London: McGraw-Hill.
Scholes, M. S., Wilson, G. P. and Wolfson, M. A. (1992) ‘Firms’ responses to
anticipated reductions in tax rates: the Tax Reform Act of 1986’, Journal of
Accounting Research, 30 (Supplement): 161–85.
 International transfer pricing and income shifting 635

Shapiro, A. (1992) Multinational Financial Management, 4th edn. Boston, MA:

Allyn & Bacon.
Sunday Times (1992) ‘Inland revenue probes tax avoidance at Sony’, 22 March, pp.
7 and 16.
Times (1994) The Times 1000 – 1994, London: Times Books.
United Kingdom Statute (1970) Taxes Management Act, 1970, section 485, now
section 770 of Income and Corporate Taxes Act, 1988.
United Nations (1978) Transnational Corporations in World Development: A Re-
examination, Commission on Transnational Corporations. Paris: UN.
Wheeler, J. E. (1988) ‘An academic look at transfer pricing in a global economy’, Tax
Notes, 4 July, pp. 87–96.
Wheeler, J. E. (1990) ‘Hearings on tax under-payments by foreign-owned US
subsidiaries’, Hearings before the Subcommittee on Ways and Means, United
States House of Representatives, April, pp. 1–15.