Content uploaded by Miklos A. Vasarhelyi
Author content
All content in this area was uploaded by Miklos A. Vasarhelyi on Apr 21, 2015
Content may be subject to copyright.
Optimizing tax allocation among
countries in the multinational entity: a
tale of many contingencies
Miklos A. Vasarhelyi
Daehyun Moon
Rutgers University
Outline
• Introduction
• International tax systems
• Linear programming application in tax planning – optimization
• Example 1: Optimal overhead allocation
• Example 2: Optimal dividend repatriation
• Conclusion
Introduction
• Globalization, deregulation, technology
– Increase in the mobility of goods and services and capital across
different jurisdictions over the world
– More aggressive tax planning is possible by multinational companies,
i.e. increased financial mobility, R&D, capital investment
• Government tax competition
– Attract foreign investment and retain tax bases and prevent tax revenue
loss from outflow of tax base
– Government induced to lower tax rates: “race to the bottom”
International Taxation
• Double tax relief
– To reduce double taxation on the same income by multiple
governments; tax treaties
• Foreign tax credit
– Many countries including US allow foreign tax credits for income taxes
paid to foreign governments
– Limited to US tax liability on foreign source income
• Tax deferral
– In US, only repatriated incomes (dividend) are taxed
– However, Subpart F (US) deems some foreign income of at least 50%
owned subsidiaries as distributed
Incentives for tax planning
• International tax policies and tax competitions in the
international environment influence multinational corporations’
business activities to avoid taxes.
• Accordingly multinational corporations engage in tax planning
to realize tax saving opportunities maximizing after-tax profits
and minimizing tax liabilities
– Income shifting
– Deferral of earnings
– Migrating activities
Income Shifting
• Two main methods of income shifting
– Locate debt in higher tax countries (thus interest expense) and invest in
equity in lower tax countries
– Transfer pricing: charge unfair prices between related parties
• Tax saving is the difference between taxes as a result of
shifting and taxes would have been levied at the origin, thus it
depends on whether the income is repatriated or kept
(Rousslang, 1997)
• However, governments like Internal Revenue Code (IRC) in
US impose rules against income shifting
– Transfer pricing rules – arm’s length prices
– Subpart F – “deemed income”, passive income
7
Deferral of earnings
• Deferral of certain foreign earning is another important tax planning
strategy –In the US, a parent firm is taxed on its subsidiaries’
income only when returned (repatriated) to the parent firm.
• The deferral of U.S taxation may create incentives for firm with
lightly taxed foreign earnings to delay repatriating dividends from
their foreign subsidiaries –the future year can be more attractive for
repatriation if domestic tax rate is lower or future sources of foreign
income generate the excess foreign tax credit that can be used to
offset the domestic (U.S) tax liability on dividends.
8
Migrating activities
• Sufficient evidence that international business will seek to maximize
profits by cost reductions under international market conditions. The
search of such cost reduction is evident in following areas
– The migration of manufacturing activities costs away from high labor cost and
high tax costs toward low-labor-cost countries.
– The shift of service activities (selling, finance, and management) away from high-
tax-cost countries toward tax haven, where tax savings in these activities can be
made.
Optimization
• Tax planning opportunities arise from many differentials
in tax rates, policies, regulations, contractual agreements
and other economy characteristics, such as labor
markets, competitions, financial market, etc. among
countries
• These different attributes should be taken account into
tax planning strategy and be optimized to maximize the
benefits of tax saving opportunities.
• Linear programming can be useful tool for solving the
optimal solutions for maximizing the potential benefits of
tax planning.
Linear Programming
• Linear programming model
– Objective function
• Maximize after-tax profit or minimize tax liability
– Decision variables
• Variables to be assigned optimizing the objective function, e.g., optimal
allocation of income to be remitted, optimal locations to invest or borrow, etc.
– Constraints function
• Conditions of decision variables, such as total investment should be no
greater than 1M, etc.
11
Example 1: Overhead Allocation
• A US based multinational corporation produces and sells
electronic items needs to allocate its overhead costs to three
foreign subsidiaries.
• The firm has business in Hong Kong (HK), Japan (JPN), and
United Kingdom (UK).
•What would be the optimal overhead allocation that
maximizes the after tax profits for all subsidiaries?
12
Examples
• Assume that the current sales levels for the each subsidiary
are :
Market Size Firms’ Market Share Tax rate
HK 1.5M 450,000 (30%) 16.5%
JPN 2.5M 375,000 (15%) 30%
UK 2M 500,000 (25%) 25%
13
Objective Functions and Constraints
• Total overhead costs to be allocated is 1M.
• The objective is to obtain optimal overhead allocation that
maximize the after tax profits of foreign subsidiaries. The
objective function is defined as
– Max [(HK sales-allocated overhead exp.)*(1-Tax_HK)+(JPN sales-
allocated overhead exp.)*(1-Tax_JPN)+(UK sales-allocate*(1-Tax_UK)]
• The objective function can be realized, subject to
– each allocated expense >= zero
– sum of optimized allocations = total overhead costs, 1M.
– for each country, allocated expenses must be less than or equal to its
net after tax sales.
Expense Allocation answer
Objective Cell (Max)
Cell Name Original Value Final Value
$B$27 Total after tax benefit 248,533.02 271,375.00
Variable Cells
Cell Name Original Value Final Value Integer
$C$30:$C$32
HK Allocated OH Exp 339,623 125,000
JPN Allocated OH Exp
283,019 375,000
UK Allocated OH Exp 377,358 500,000
Constraints
Cell Name Cell Value Formula Status Slack
$C$33 Allocated OH Exp 1000000 $C$33=$E$27 Binding 0
$C$30 HK Allocated OH Exp 125,000 $C$30<=$B$30 Not Binding 325000
$C$30 HK Allocated OH Exp 125,000 $C$30>=0 Not Binding 125,000
$C$31 JPN Allocated OH Exp 375,000 $C$31<=$B$31 Binding 0
$C$31 JPN Allocated OH Exp 375,000 $C$31>=0 Not Binding 375,000
$C$32 UK Allocated OH Exp 500,000 $C$32<=$B$32 Binding 0
$C$32 UK Allocated OH Exp 500,000 $C$32>=0 Not Binding 500,000
15
Optimized Overhead Allocation
• Given constraints and objective function, the Linear
Programming model finds optimal value of decision
variables of objective function.
• The maximized value of the objective function which is
total after tax profits of subsidiaries in this example is
271,375 and optimized decision variables (overhead cost
allocations) are determined.
HK JPN UK Total
Optimal
Allocation
125,000
(13%)
375,000
(38%)
500,000
(50%)
1M
16
Example 2: Dividend Repatriation
• For this example, consider the creditors of US parent
corporation require the dividends from 3% of the total
foreign operating assets from the foreign subsidiaries.
• Assume that the total foreign assets are 20M and the
dividends to be remitted for total 600,000 according to
the contractual agreement
•What would be the optimal way to allocate the
dividends to be remitted and to minimize the US tax
liability on the dividend incomes from the
subsidiaries?
17
Objective Function and Constraints
• Assume that all other conditions are the same as the prior example.
• The objective is to minimize the US tax liability on dividends and decision
variables to be optimized are the amount of dividends remitted by each
subsidiary. The objective function would be
– Min [((US before tax sales) + (Dividends from HK + Dividends from JPN
+ Dividends from UK)) * Tax_US) - (Dividends from HK * Tax_HK) -
(Dividends from JPN * Tax_JPN) - (Dividends from UK * Tax_UK)]
• Subject to
– Total of optimal dividends amounts = 3% of total foreign assets which is
20M
– Each subsidiary’s dividends remission <= its after tax sales
– Each subsidiary's dividend remission >= zero
Dividend Repatriation answer
Objective Cell (Min)
Cell Name Original Value Final Value
$I$29 Total Tax Liability 364,143 342,188
Variable Cells
Cell Name Original Value Final Value Integer
$E$31:$E$33
HK Div. 203,773 0
JPN Div. 169,811 262,500
UK Div. 226,415 337,500
Constraints
Cell Name Cell Value Formula Status Slack
$E$34 Div. 600,000 $E$34=$C$27 Binding 0
$E$31 HK Div. - $E$31<=$D$31 Not Binding 375750
$E$31 HK Div. - $E$31>=0 Binding -
$E$32 JPN Div. 262,500 $E$32<=$D$32 Binding 0
$E$32 JPN Div. 262,500 $E$32>=0 Not Binding 262,500
$E$33 UK Div. 337,500 $E$33<=$D$33 Not Binding 37500
$E$33 UK Div. 337,500 $E$33>=0 Not Binding 337,500
19
Optimized Dividends Repatriation
• The objective value, minimal value of total US tax liability
is 342,188, and its optimal dividends repatriation of each
subsidiary variables are determined by Linear
optimization model.
HK JPN UK Total
Dividends
To be remitted
0
(0%)
262,500
(44%)
337,500
(56%)
600,000
(100%)
20
Conclusion
• Linear Programming (LP) is used to find optimal decision variables
that maximize the objective functions.
• Finding optimal solutions for various tax related decisions are critical
to achieving innovative tax planning to reduce tax liability and
maximize the after-tax benefit.
• While examples are purposely made simple to show the general
concept of LP and its application on international tax planning, the
linear objective model can solve the hundreds of decision variables
with thousands of constrains functions.
• As the complexity of tax planning increases, the role of LP and other
computer-based expert systems will be effective and powerful tool
for providing successful tax planning services.
21
Future Research
• Linear Programming (LP) and other computer-based
expert system or mathematical operation research
methods have been used in solving problems for many
areas such as biochemistry, medicine, computer
science, etc.
• However, LP application in international tax planning
field is rarely found, although normative researches on
optimal tax policy often use optimization approach.
• Other than LP, tax planning can also be analyzed with
structuring decision trees where the ramifications of
strategic decisions are made to follow a critical path that
minimizes tax costs and maximize the post tax profits.
22
Bibliography
• Michael W.E Glautier and Frederick W. Bassinger, “A
Reference Guide to International Taxation”, Lexington Books,
1987.
• James R. Hines,Jr. “Taxes, Technology Transfer, and the
R&D Activities of Multinational Firms”, The Effects of Taxation
on Multinational Corporations, The University of Chicago
Press, 1995.
• Kenneth A. Froot and James R. Hines, JR., “Interest
Allocation Rules, Financing Patterns, and the operations of
US Multinationals”, Working Paper No. 4924, National Bureau
of Economic Research, 1994.
• Quing Hong and Michael Smart, “In Praise of Tax Heavens:
International Tax Planning and Foreign Direct Investment”
CESifo Working paper No. 1942, 2007
