My parental leave has officially started and I have limited time for blogging (or sleeping).  Friend and WashU law faculty member Adam Rosenzweig has written up another guest post on how firms can avoid (or minimize) corporate taxes.  See below.

___________________________

Imagine you are a hedge fund in the business of investing in high-yield bonds (otherwise known as junk debt).  Investors in the hedge fund come from all over the world, including US individuals, foreign individuals, corporations and pension plans.  An opportunity arises to make a loan to a US company as part of a leveraged buyout of the company.  You are extremely interested in making the loan.  Should you do it?

Of course, the answer ultimately is a business decision.  But what if I told you that the LBO loan would always have a thirty-five percent lower return than junk bonds purchased on the market?  How could this be the case?  Taxes.

For the most part, hedge funds are organized offshore so as not to be subject to US tax.  This is primarily for the benefit of their foreign investors who typically are not subject to US tax themselves and thus do not want to invest in entities that would be subject to US tax.  The problem is that foreign entities investing in US assets can be subject to US tax if they look too much like US business.

More specifically, under US international tax rules non-US persons (such as offshore hedge funds) are not subject to tax on their income unless that income is either, (1) fixed, determinable, annual or periodic income (“FDAP”) paid from sources within the United States, or (2) effectively connected with the conduct of a US trade or business (“ECI”).  The first refers to payments such as interest and dividends.  Such payments made by a US corporation to a non-US person are subject to a thirty-percent gross withholding tax (unless a treaty applies to reduce the rate).  Hedge funds can relatively easily avoid this provision, either by buying assets that do not pay FDAP type income or by entering into derivatives against the assets paying FDAP income instead of owning them directly (subject, as always, to certain anti-abuse rules).

The second refers to non-US persons engaged in a trade or business in the United States (or if a treaty applies have a permanent establishment in the United States).  This rule treats all income effectively connected with such a business as if it were earned by a US business.  The idea is to put US and non-US companies competing in the United States on equal footing.  This ECI is subject to a net income tax (meaning income less deductions) under the same rates as US people.

The problem arises because investing one’s own money is not treated as a trade or business for these purposes but banking is.  The tax law has a hard time figuring out which is which, however.  An entity need not take deposits and make individual loans to be in the banking business.  Rather, any entity that advances money and seeks out borrowers for the money (or “originates” the loan) can potentially be in the banking business for US tax purposes.  So if the hedge fund seeks out the US company, negotiates the terms of the loan, and advances the money, it sure looks like it is in the US trade or business of banking.  If so, all interest on the loan would be subject to US tax.

The solution is for the hedge fund not to originate the loan but instead to passively buy and sell an existing loan (for bank loans this is accomplished by buying “participations” in the loan).  But if the hedge fund truly was passive and waited to see what loans came on the market it would not be able to control the terms of the deal in the first place, meaning the payout may not be as good as the fund expected when it found the business opportunity.

The “season-and-sell” solves this problem.  Under the season-and-sell, a hedge fund identifies a lending opportunity in the United States.  It then forwards the opportunity to a friendly US bank with the proposed terms.  The US bank negotiates and makes the loan.  Then, the US bank waits for a period of time (how long is an issue of contention, discussed below) and then sells a participation in the loan to the hedge fund.  Assuming each step is respected, it now looks like the US bank originated the loan and that the hedge fund merely purchased a passive investment in debt.

Voila – the hedge fund gets the investment it wants, at the terms it wants, but is not engaged in a US trade or business.  Yes, the US bank will charge a fee for this service, but that is much less than the US tax would have been (and also avoids all the messy US tax return filings).

The season-and-sell has proven both popular and controversial.  Some people find it offensive that by using a US bank as a “front” and merely waiting to buy into the deal a hedge fund can completely avoid US tax.  Such people contend that the entire transaction should be collapsed into a single hypothetical transaction since the intervening steps are hard-wired and in substance the hedge fund made the original loan.  Under this hypothetical, the hedge fund would be treated as directly lending to the US borrower and thus would be engaged in a US trade or business for tax purposes.

Proponents of the season-and-sell anticipated this argument, however.  By waiting a certain amount of time before selling the participation to the hedge fund, they contend that the US bank takes on real risk.  If the price of debt drops in that period of time the bank loses real money.  Although it is unlikely that the price of debt would drop that quickly, those around in 2008 can easily remember seeing the value of mortgage loans crashing in a matter of days.  So how could an investment where the US bank takes on real risk of loss be a sham and disregarded for tax purposes?  And if the tax law did pretend like it never occurred, what would happen if the bank really did lose money?  How would the tax law account for that?

Based on these arguments, the market seems comfortable that season-and-sell works for tax purposes so long as the period of time is long enough.  How long would you guess is long enough?  One year?  One month?  As Lee Sheppard, a leading tax journalist, noted “[w]e’re not talking fine wine here. We’re not even talking Beaujolais Nouveau.”  Although there is no clear answer, the practice seems to be that three days is enough.  Yup, three days.  By paying a US bank a fee and waiting three days, the theory goes, the hedge fund can make any US loan it wants without paying US tax.

I am ambivalent about whether season-and-sell works under current law.  More interesting to me, however, season-and-sell demonstrates pretty starkly the limits of the existing US international tax laws to modern finance.  The US corporate tax system was designed to tax real multinational companies selling real stuff like General Motors or Standard Oil.  It is debatable how effectively it does so in the modern world, but season-and-sell seems to make clear that rules meant for companies that sell real stuff are toothless in the face of financial businesses.  After all, to a hedge fund a dollar is a dollar.  They don’t care whether an investment is called a loan, or a participation, a bond, a total-return swap, or anything else, so long as the cash flows are the same under their models.  So they can buy and sell any of these, or turn one into the other, solely for tax savings.  By contrast, I am assuming it would matter to General Motors whether they sold a Chevy Corvette or Ford Mustang, notwithstanding that both are cars.

As a result, it is nearly impossible to apply a substance-over-form analysis to a business model (such as a hedge fund) that is completely indifferent to form.  In other words, when is a participation in “substance” a loan and in “substance” a passive investment when to the hedge fund they are completely identical?  That, ultimately, is the true weakness of the US international tax rules.  Season-and-sell is just one (relatively clever) example.