Leveraged Long/Short Tax Alpha Is a Hotel California

​The Spreadsheet That Ruins the Afternoon

Every time something becomes the hottest trade on Wall Street, someone, somewhere, sits down with a spreadsheet and works out the actual unit economics over time, and posts the spreadsheet on the internet, and ruins everybody's afternoon.

The person who did that here is Victor Haghani at Elm Wealth, whose paper “Tax Aware Long Short,” cited by Matt Levine in Money Stuff, which is how we know something is a big deal, concludes that the manager's fees over twenty years eat up roughly three-quarters of the capital gains tax the investor was trying to defer in the first place.

But the bigger problem is that after nearly two decades of harvest cycles, there are highly leveraged long positions that need to be unwound and deeply in-the-money shorts that have to be covered.

"The investor has, in the technical language of the literature, ossified," says Mike Allison, Strategist and Portfolio manager at Investment Research Partners and former Global Portfolio Manager and Director of Equity Strategy Implementation at Eaton Vance.

Unwinding the longs at that point triggers long-term capital gain at roughly 24 to 29 percent all in.

"Unwinding the shorts is worse," Allison says, "because Section 1233 says that closing a short sale is always taxed at short-term rates, which in a state like California means roughly 54 percent."

So now you have a portfolio you cannot unwind without writing that big check to the IRS you were hoping to avoid.

What do you do?

You might be up on tax trends and conclude that the answer is a Section 351 exchange.

For background, Section 351 is the corner of the tax code that lets you contribute property to a new corporation in exchange for its stock (or ETF shares) and defer the gains.

For long-only direct-indexing SMAs, this has worked beautifully, asset managers like Cambria, Alpha Architect, and Polen have been rolling up hundreds of millions of dollars of locked-up legacy positions into newly seeded ETFs.

Investors own ETF shares with a carryover basis, the ETF owns a clean, tax-efficient portfolio; everyone goes home happy.

I would love to tell you the trick works on tax-aware leveraged long/short portfolios. It does not.

A Short Position Is Not a Thing You Can Contribute

The reason it does not work comes down to what a short position legally is.

When you short a stock, you have not bought anything, you have borrowed someone else's stock, sold it, and parked the cash proceeds in a margin account.

Revenue Ruling 95-45 says the obligation to return the borrowed stock is a liability, and the only "asset" the short produces is the cash sitting in the margin account.

A Section 351 exchange requires each contributor to bring a diversified basket of securities, not liabilities, that independently meets the RIC diversification test at the contributor level, and cash does not count toward diversification.

You might then think, fine, leave the shorts at the SMA and just 351 the long book.

Except the longs in a 250/150 portfolio are not really owned in the way Section 351 requires.

You are long $250 on $100 of equity. The other $150 is borrowed, secured by the long positions themselves.

Margin debt is a liability.

You cannot contribute pledged collateral, and you cannot contribute the longs together with the margin debt either.

The leverage that manufactured the tax losses on the way in is the liability that traps the investor on the way out.

Getting desperate, someone might ask, ‘What if the receiving ETF is itself a long/short fund, like a market-neutral ETF? Couldn't it just take the shorts in kind, since it is going to hold shorts anyway?’

No. What the receiving fund plans to do is irrelevant.

What matters is what a short position is in the eyes of the Internal Revenue Code, which is a liability, regardless of who ends up holding it.

A long/short ETF can build its own short book from scratch using its own cash after a Section 351 exchange of long positions only. But it cannot take your shorts in kind.

Two Workarounds, Both Bad

So that is the situation. Hundreds of billions of dollars are pouring into tax-aware long/short funds that cannot, as a structural matter, be 351-exchanged out the back end.

There might be two workarounds, but sorry to report that they are also bad.

Option 1: The IRS recently launched a fast track private letter ruling program that, in theory, makes asking for the same blessing easier than it used to be. In practice, you would still need a custodian willing to transfer margin balances and short borrow lines into a brand new ETF, which none are built to do. If you could find one and a sponsor willing to pay for the ruling, it is technically not impossible. Realistically, this is not happening.

Option 2: Skip the corporation and contribute the portfolio to a partnership under Section 721 instead. Partnerships have a much more permissive liability netting framework, and as long as the partner stays meaningfully in the partnership, the short liability generally washes out. The IRS has blessed this, and the Tax Court has upheld it. The catch is the mixing bowl rules, which trigger gain to the original contributor if the partnership distributes the property to anyone else within seven years. So, you can defer the tax, but you cannot diversify out for seven years, which is a long time to wait if your whole thesis was that you wanted to be diversified. The partnership could, in theory, then 351 the long positions into an ETF, which is roughly what Cache, the new exchange fund startup, has been promoting. But the partnership still has to distribute the capital gains on the short positions, which means the short problem never actually goes away. It just moves.

The Hotel California of Tax Planning

​It is worth saying that this is a narrow problem, and drawing the distinction between these new leveraged funds and the much larger universe of long only direct indexing and tax loss harvesting strategies that keep a long only book. The Parametric type strategies that now run close to a trillion dollars, and are still growing, have none of these issues. Long only baskets contribute cleanly into Section 351 ETFs.

The tax problem, ironically, is specifically with the leveraged long/short version, the newer, sexier, more sophisticated cousin, that does not work when investors want out. For the wealth advisors and asset managers and high net worth families pouring in the capital, it is the Hotel California of tax planning. Usually someone on Wall Street eventually comes up with a clever workaround. This one does not look promising.

Disclaimer

Please note this article is for information purposes only and does not in any way constitute investment advice. It is essential that you seek advice from a registered financial professional prior to making any investment decision.