Transfer Pricing: Financial Models and the Application of CAPM in Financing Arrangements
Article by: Argyris Emmanuil, Manager Tax Services – Transfer Pricing Services, Baker Tilly South East Europe
Nowadays, transfer pricing documentation faces several challenges, as transactions between related companies are becoming increasingly complicated and Tax Authorities increase their documentation requirements with regards to the presentation and analysis of such transactions.
In many cases, Transfer Pricing experts use different models to approach and analyse the controlled transactions. Since transfer pricing is not considered an absolute science, documentation methods may highly diverge among experts and in many cases be questioned.
In the case of financing arrangements, certain industries rely on economic models to price intra-group loans/financing arrangements by constructing an interest rate as a proxy to an arm’s length interest rate. In their most common variation, economic models calculate an interest rate through a combination of a risk-free interest rate and a number of premiums associated with different aspects of the loan – e.g. default risk, liquidity risk, expected inflation or maturity (p.10.105).
As per OECD guidelines:
10.106. The reliability of economic models’ outcomes depends upon the parameters factored into the specific model and the underlying assumptions adopted. In evaluating the reliability of economic models as an approach to pricing intra-group loans it is important to note that economic models’ outcomes do not represent actual transactions between independent parties and that, therefore, comparability adjustments would be likely required. However, in situations where reliable comparable uncontrolled transactions cannot be identified, economic models may represent tools that can be usefully applied in identifying an arm’s length price for intra-group loans.
In transfer pricing reports, financial models consist of a usual practice to document different types of transactions. The Capital Asset Pricing Model (CAPM) is one of the most widely used models for estimating such returns in the investment world.
For instance, a TP expert applies CAPM to capture the higher risk resulting from receivable loans and financing arrangements which are being financed out of third-party funds. CAPM is usually presented as a risk premium in addition to the initial remuneration (handling fee).
Alternatively, receivable loans financed out of equity could be documented by the CUP method under a benchmarking interest rate analysis, while back-to-back loans among related parties where a Company only assumes limited risks and performs limited functions, a handling fee analysis should be well accepted as a documentation method.
Additionally, Financial Models are also suggested in the newly presented chapter X of OECD guidelines and more specifically in the documentation of financial guarantees.
However, several challenges of the CAPM model were raised in the financial years of 2020 to 2022 which would highly suggest that the CAPM model should be adjusted or replaced by better alternatives.
The effects of COVID-19 in combination with other geo-political events that took place during the last years, resulted in a spiral of economic shocks. The unanticipated financial shocks of the last 3 years have been characterized as one of the worst episodes of global financial distress in decades, while the results show that contagion was systemic during this period. As a result, the CAPM and other economic models have been highly questioned.
Therefore, CAPM illustrates the relationship between expected return and systemic risk as follows:
Expected Return on Equity = Risk-free rate + Beta x Risk Premium
Re=Rf + β x (Rm – Rf) Re=Rf + β x Pc
There are many opinions on the failure of CAPM, challenging each of its components such as the Beta, the Risk Premium and the Risk-Free rate.
However, due to the above-mentioned socio-economic events, the failure of CAPM can be mainly justified in the calculation process of free risk rate and how is being approached.
There are two types of risk-free rates: the nominal risk-free rate and the real risk-free rate. The difference is due to the impact of inflation.
The nominal risk-free rate is typically the current yield of the 3-month T-bill without taking into account the impact of inflation. The real risk-free rate is the yield of the 3-month T-bill minus the impact of inflation. It is at the discretion of each Financial analyst if the US T bill would be used as a general practice or if a domestic bond would be a better proxy as so.
The real risk-free rate, on the other hand, is the yield an investor would need on a prospective investment so as not to experience inflation risk – providing inflation rates stay the same or decrease.
U.S. Treasury bills, commonly referred to as T-bills, offer the shortest maturity debt securities issued by the federal government. They are the closest that investors can get to a zero-risk investment, for two reasons:
Firstly, as noted above, there is a nearly universal belief that the U.S. government will always make good on its debt. Unlike corporate bonds or municipal bonds, T-bills are backed by the full faith and credit of the U.S. government, which has never defaulted.
Secondly, the market for U.S. government debt is the largest and most liquid market anywhere in the world. As of April 2022, SIMFA reports that the market cap for U.S. Treasury-issued securities was $23.3 trillion, with $679 billion being actively traded.
Real Risk-Free Rate = (1+10 years government bonds rate[1]) / (1+inflation rate[2]) – 1
Real free risk rate in the last 3 three years mainly fails in emerging markets due to inequalities in their government bonds and inflation.
For instance, in 2022 the Government bond of Bulgaria was 4,297% while the Inflation rate was 13,02%. Therefore, by investing in Bulgaria’s government bond you’d be falling 8,723% short of keeping pace with current inflation rates.
The above argument is also supported by the TP guidelines, where turning to the methodologies, the availability of CUPs is discussed in paragraphs 10.89 to 10.95. Paragraph 10.93 suggests that comparables for an intra-group loan could include bond issuances, loans which are uncontrolled transactions, deposits, convertible debentures, and commercial papers.
Guidelines suggest that deposits could be comparable to loans but considering the unsustainable economic environment in 2019 to 2022, negative deposit rates made their appearance.
Could negative deposit rates be compared to loan rates?
Since the answer is no, an investor could examine the opportunity cost of each of its investment instruments. The lender’s perspective in the decision of whether to make a loan, how much to lend, and on what terms, will involve evaluation of various factors and possible best alternatives out in the market, such as investment in bonds, deposits, securities, and other derivatives.
If your goal is to grow your money and retain purchasing power, you’ll probably want to look for a different investment with higher yields. In an inflationary economy, that typically means a riskier investment.
Going back to transfer pricing methodology if a TP expert wants to keep using CAPM then this should be adjusted in order to represent the reality of the economy.
One way is to set the inflation rate as a target by setting an additional notional rate, and the second possible measure could be to only consider the nominal risk-free rate which is equal to the government bond of the relevant economy or equal to the US Treasury bill.
Finally, chapter X of TPG suggests alternative models which a Transfer Pricing expert should assess in order to identify the most suitable documentation method with regards to risk premiums of financial loans or in estimating the rate of financial guarantees.
TPG suggests that the risk premium associated with intra-group loans could also be calculated by the spreads of credit default swaps (p. 10.101) bearing several advantages and disadvantages.
Financial guarantees could be estimated by applying the Yield Approach which quantifies the benefit that the guaranteed party receives from the guarantee in terms of lower interest rates; or
by the Capital Support method which takes into account the borrowers’ and the guarantors’ credit ratings; the difference among the two credit ratings identifying the additional notional capital to the borrower’s balance sheet to bring the borrower up to the credit rating of the guarantor. (p.10.174 to 10.182). Such a functional analysis is likely to include consideration of similar information to which a commercial lender or ratings agency would consider in determining the creditworthiness of the borrower.
Concluding, each transaction is different and has its own characteristics. In each case, one needs to examine the economic environment, the market environment, collaterals, and its terms and conditions, etc (p.10.28 & 10.29).
Alternatively, financial analysts and transfer pricing experts could test other financial models along with the CAPM such as the Arbitrage Pricing Theory (APT) model. The Arbitrage Pricing Theory (APT) is an alternative to CAPM. With APT, each asset’s payoff will come out as a weighted average of all the rest in a portfolio. The APT formula is:
E(rj) = rf + bj1RP1 +bj2 RP2 +……bjnRPn
The idea behind APT is that an asset’s return depends on two key factors: the macroeconomic environment (inflation, interest rate fluctuations, etc.) and the possibility that the asset will move according to environmental factors.
Additionally, one could consider the Multifactor model, which uses historical data to relate stock returns to specific macro-economic variables (the level of interest rates, the slope of the yield curve, and growth rate in the GDP) and estimates betas for individual companies against these macrofactors.
Both models represent extensions of the CAPM, with multiple betas replacing a single market beta, with risk premiums to go with each one. These models are much more complicated but considering the overall macroeconomic effects with multiple components such as beta can eliminate the shortcoming of a possible failure of a single component.
As per OECD guidelines:
10.28. There is a wide variety of financial instruments in the open market that present very different features and attributes, which may affect the pricing of those products or services. Consequently, when pricing-controlled transactions, it is important to document the transactions’ features and attributes.
Finally, various approaches exist which may succeed or fail from time to time to present the true image of the market. At the end of the day, an analyst needs to present rational conclusions resulting in a rational documentation result. Assumptions and hypothesis applied could be generally accepted as long as they are rational and they do not diverge from the OECD guidelines.
