I.R.S. Cracks Down on Hedge Fund Tax Strategy
nytimes.com
nytimes.com
http://www.bloomberg.com/news/articles/2014-07-21/renaissanc...
Essentially the US has a few different taxes you can pay as an investor. There is a short term tax on trading profits and a longer term holder tax for "investors" which is lower.
This obviously creates a situation where people will do their best to make their trading fall into the later category.
What Deutsche did was to sell hedge funds an option on a basket(collection) of stocks that was held longer than a year. Not surprisingly, a year is long enough to qualify for hte lower long term investment tax on capital gains.
The IRS has called shenanigans on this and the below quote is probably the biggest reason why:
> "Illustrating how rapidly the contents of the “baskets” were shuffled, one option reviewed by the committee had more than 129 million underlying trades in a single year, the subcommittee said. Many of Renaissance’s stock investments lasted mere minutes or seconds, it said."
I buy stock in a firm, that firm makes a lot of profit, but I don't liquidate my position. How is that different (short-term vs long-term gains) from me selling their stock and buying another stock?
It's even the same word: A retail business sells its stock (product), and sells it's stock (shares)
Ultimately, taxing "realized" gains can never be logical, since "realization" is a fictional concept not grounded in reality. Taxing consumption is relatively straighforward to describe (but hard to implement), and taxing wealth is slightly harder (since it's hard to mark the value of some assets)
Today's step function, with made-up tax percentages: 20% if held for less than a year, 10% if a year or more. You end up with a bunch of trades held for a year and a minute.
My proposal, with interpolation: 20% if held for less than 6 months 10% if held for 18 months or more Between 6 and 18 months, we interpolate between 20 and 10%.
The difference between waiting 364 or 365 days is now minimum. I chose 6 and 18 because they're 12 +/- 6. However, a different curve could be drawn. The idea is to replace the step with a gradual increase.
I don't think this level of extra computational burden is a good counter-argument against more sane tax policies in this day and age.
I use an intermediary, but have to report individual stocks nonetheless (fraud avoidance?) And you can only require what they can provide. During tax season, I have to dive through filing cabinets to fish out the cost basis information that's missing on older investments made before they were required[1] to start recording that information, plus double checking splits online to make sure this wasn't a partial sale that I have to calculate the cost basis on. They report what they have on Form 1099-B.
Right now you have to file a separate Form 8949 for each combination of: (Short Term, Long Term) x (Reported to IRS & Me via 1099-B, Reported to only Me via 1099-B, Unreported to Me), plus Schedule D, plus maybe Schedule B. Last year that was "only" five pages of tax forms filed for two stock transactions + various dividends, plus a few worksheets that you don't file. I forget if there was a cabinet dive.
Given how obtuse tax forms get, a tax-rate lookup for each transaction would probably involve a worksheet per stock, just to do the date time calculation of "how long was this stock held".
[1] At least, I think I read they were required to start tracking this.
Maybe some stock brokers or weird investments don't do this?
I would hate having to do all that stuff by hand.
I have a couple small positions I've held for more than 20 years now. At this point it will be more trouble to deal with the taxes than they're worth, so I'll probably just hold onto them and let my heirs deal with it.
How is this not "logical" ? Can you explain better?
See other comments in this subthread that illustrate the various ways to structure assets.
Trading an asset vs. holding it is a non-fictional concept very much grounded in reality.
Attributing significance to it is, in a sense, arbitrary in the same sense that any assignment of significance is, but its definitely not a fictional concept divorced from reality.
What exactly are you trading? Bits in a database that mark ownership of a fraction of a fictional entity?
Nope (although, that may be true, in some sense), I am not saying that. I am saying the thing many refer to as an 'entity', as in, 'the corporate entity', is a fictional thing. It has no real, inherent or independent physical existence apart from the people that choose to represent it.
Please read more carefully. I never claimed any such thing.
'fictional' modifies 'entity', not 'ownership'.
Will need to rethink my involvement here.
Part ownership of a company.
To say anything else is to show yourself to be completely naive.
To believe that all assets are 'part ownership of a company' is not only naive, but incorrect and delusional.
But $10 and ten $1 bills are equivalent. A tax that applied to one and not the other would be very weird; it's not clear what behavior it's meant to encourage, and anyway lots of people would avoid paying it.
So the argument above is that "realizing" gains is a lot like getting change for a dollar bill. I don't know whether that's right, but anyway it's different from arguing about whether money is real :)
1- Over the course of a year I buy and sell hundres of stock issues, constantly incuring taxable events, and at the end I sell off the entire portfolio including short term cap gains rate (same as income rates).
2- I buy an ETF that does all the same trades, but when I sell I only get long term cap gains taxes.
3- I loan a company money (sell an note) to do this trading and a clause written for at the end of the year call the note for a value equal to the previous two scenarios. I incure taxation from a bond investment,
4- I buy an ETN (exchange traded note) that repays based on a formula as if the previous trades were made. Once again I pay long term capital gains on an equity investment.
5- I buy a future on the index I'm managing and invest an an amount equal to current price of the stocks in bonds to replicate the price performance on the ETN. When I sell a year, I incure two taxable events.
6- I have't even brought up options, so I can keep going.
These are all the same, but they all have different taxable events and consider a realized gain something different.
“A retail business (...) sells it's stock”
The contraction “it’s” means either “it has” or “it is”. So, the quoted expression expands either way to a syntactically malformed expression with two verbs.
Could it be intended to mean the following?
A retail business’ (...) sells — it has stock U.S. Constitution, Article I, Section 9:
No Bill of Attainder or ex post facto Law shall be passed.
"The new I.R.S. guidance will be retroactive, applying to all transactions as far back as Jan. 1, 2011."
?
Not to mention that there is solid precedent that retroactive taxes are not, in general, ex post facto laws in the first place, such that even if this was Congress adopting a retroactive tax law, it still wouldn't fall afoul of the ex post facto prohibition.
> IRS actions under the current interpretation, the law which
> would be controlling is not retroactive.
This is to me a very weak argument (not that I disagree that this may be the standard that is applied in practice). If the Congress passed a law saying "the IRS can set tax code as it sees fit" and the IRS proceeds to levy taxes all the way back to 1980 on people as however it so desires, you should not be allowed to use that law as grounds for not having violated the ex post facto requirement. If this is the current understanding of what is allowed, I must vigorously object to whichever judgement left this precedent.
Then this would be an unconstitutional delegation of legislative authority to the executive. [0] Rendering the consideration of whether any executive action under it would constitute an ex post facto law moot -- if Congress purports to delegate enough power to the executive that retroactive executive action under it would be an ex post facto law, than the whole scheme is an unconstitutional delegation of legislative power whether or not it is applied retroactively.
Heck, even though retroactive taxation isn't generally ex post facto law (which is retroactive criminalization or enhancement of criminal penalties), the law you propose would be an unconstitutional delegation.
> If this is the current understanding of what is allowed
Its not, its just the prohibition that prevents it has nothing to do with (and is much broader than) the ex post facto law prohibition.
[0] See, e.g., https://en.wikipedia.org/wiki/Nondelegation_doctrine#United_...
And yet this is what we seem to be experiencing in this case, where the IRS can make a policy interpretation retroactively to the tune of $6 billion for a single taxpayer. Your source indicates that congress allows the IRS to decide tax policy, but since tax evasion is itself a crime it seems to indirectly violate the ex post facto law prohibition.
No, Congress passing a law saying that "IRS can set taxes at any level they want" is not what we are seeing.
If you want to argue that the actual laws Congress has passed are unconstitutional delegations, please, point to the specific laws, and make that argument.
> but since tax evasion is itself a crime it seems to indirectly violate the ex post facto law prohibition.
Evasion is a different thing than non-payment. No act that occurred in the past that was not evasion when it occurred becomes evasion as a result of this change in application.
Incorrect, but I can see how you might think so. It is the same because if I disagree with the IRS and refuse payment then according to https://www.law.cornell.edu/wex/tax_evasion, I have committed tax evasion.
So yes, since the IRS can adjust my tax burden retroactively, tax evasion is involved. Suppose in 2012 I filed my tax returns and was fully honest. The IRS agreed, but now in 2015 they change the 'interpretation', and if I disagree with the new tax they are levying on me, presto chango! I can now be convicted of tax evasion for refusing to pay this arbitrary retroactive tax.
Nonpayment and refusing payment are different things. The act of refusal would happen after the policy change, therefore, the only act that might even arguably be criminalized is an act that would occur after the change, and no retroactive criminalization has occurred.
No act before the policy was adopted is criminalized, even by your characterization.
Now, if the actual law allows arbitrary changes to tax calculation by the IRS, you could argue a nondelegation doctrine violation, rather than an ex post facto violation. But to argue that, you'd actually have to point to the provision of law at issue.
Or you could argue that the policy change is inconsistent with the law Congress has adopted. But, again, you'd have to actually point to the specific provisions that are violated to do that.
If, finally, the IRS simply failed to correctly apply the law as Congress wrote it previously, and the change in guidance is simply closing a gap between IRS policy and the law written by Congress, then you can maybe fault what the IRS was doing before (though, of course, no taxpayer would challenge an overly lax application of the law -- or even have standing to do so), but not what they are doing now.
The NYT even called it a retroactive policy change!
You can ignore whatever you want, but it doesn't change what the words in the Constitution meant in the context they were written, and have been consistently been interpreted by the courts to mean. Arguing that something is unconstitutional just because you've invented an entirely ahistorical set of definitions of the words used is, well, not all that unusual, but still not particularly interesting.
> It doesn't have to be as complex as you make it out to be.
Its not at all complex: A government action is an ex post facto law if it criminalizes, or increases the criminal sanction for, an event that occurred before the act occurred. Even by your own characterization, the only thing this might criminalize is a refusal to pay the newly-calculated tax after the policy. So, no ex post facto law.
You can argue that it is undesirable for other reasons, but you can rest on the Constitutional prohibition of ex post facto laws to do that when its not an ex post facto law.
> The NYT even called it a retroactive policy change!
Its obviously a retroactive policy change, in that it applies to IRS assessment of taxes for prior tax years.
No act that was committed in those tax years becomes criminal because of the policy change, so its not an action which retroactively criminalizes an act or increases the criminal penalty for an act.
"Retroactive policy change" and "ex post facto law" aren't the same thing.
I guess people have different definitions of evasion, but I would probably define it as having the intent to evade taxes "from the start". As a citizen (or a corporation), you are free to decide on your own interpretation of the law. Tax authorities (around the globe) aren't always right. Hence you ask a judge, because an interpretation is just that: an interpretation. It's not the law. And the law is the only thing that counts.
The legal standard for proving tax evasion disagrees with you.
Innocent until proven guilty?
Yes, for refusing to pay. If you concede to the IRS interpretation, you just pay the amount you owe plus interest and fees.
> this arbitrary retroactive tax.
that's not what it is
"Battery" is a crime. If the state decides that "smacking someone's back" meets the definition of battery, they can prosecute people who committed that act before the clarification, as long as a judge finds their interpretation to be consistent with the wording of the law.
Law is not computer code that fully specifies a simulation in advance.
The parallels to the IRS case should be obvious, and the above hypothetical scenario is very much what the constitution was meant to protect against.
Please point the specific provision(s) of law parallel to the one in your hypothetical, both as to the unlimited discretion and as to the discretion to decide what constitutes criminal punishment.
Because I don't think the parallels are at all obvious.
I never said the discretion was unlimited; rather, I think that any retroactive decision in guidance/policy/interpretation is prohibited by the constitution. The words are just semantics.
IRS tax avoidance constitutes criminal punishment.
Thanks for the downvotes, everyone.
You keep asserting that, but you have failed to even begin to argue that.
> rather, I think that any retroactive decision in guidance/policy/interpretation is prohibited by the constitution.
You are welcome to think that, however, that's not consistent with the history of the use "ex post facto law", which was a legal term of art already in existence at the time the Constitution was written, and whose scope of meaning in the Constitution since, practically, the ink was still wet on the Constitution.
> IRS tax avoidance constitutes criminal punishment.
No, tax avoidance (which is structuring activities in awareness of the law so as not to incur tax liability) isn't something that is subject to criminal punishment; it is perfectly legal.
Nonpayment of tax isn't legal, but isn't generally criminal either.
Tax evasion is criminal, but involves things like false representation to avoid tax liability being correctly assessed. Following the then-current IRS guidance isn't evasion, and doesn't retroactively become evasion when the guidance changes, even though the change in guidance may change the IRS's computation of tax due.
There seems to be no good word for nonpayment of tax; tax evasion you take issue with elsewhere, tax avoidance is not the correct word I agree, but I don't know what word you really would like.
One thing is very clear: the IRS will take what it feels is due ('legally', or retroactively legal) and is willing to use the criminal justice system to do so. I and any freedom-loving American see this as a gross violation of the constitution.
I consider the IRS as a necessary and permitted part of governance, as a way to pay for the many benefits I enjoy for living in a (mostly) civilized nation.
I at least of course do not see the IRS proper as a gross violation of the constitution. Rather, the retroactive judgement of tax owed is a gross violation. That is the context we are having this conversation in, whether the government can retroactively decide that you owe tax to it, in apparent violation of (the spirit, anyway) the constitution.
In other words, the law is and aways has been that one should pay the appropriate taxes. OTOH, guidance informs the IRS how to interpret market activity in the past and classify it as legal or illegal.
Retroactive tax laws are not ex post facto laws, and, anyway, this isn't even a retroactive tax law, its a decision about the manner in which the IRS will apply a law already on the books, which is only sustainable to the extent to which it is, in fact, consistent with the law already on the books. Its essentially the IRS coming to the view that they have not been properly enforcing existing law, and that will change for enforcement going forward (there are limits to how far back the IRS can go in enforcement actions, which presumably is the basis for the 1/1/2011 date.)
...but that justification seems like a horribly slippery slope. If you accept that logic, what stops congress from passing a law that creates the "Criminal Investigation Service" and gives their director the power to make things illegal? Would he then be allowed to make a "policy change" retroactively making something illegal and arresting them for it?
edit: I think Peugh v. United States actually answers this. Retroactive application of changes to US sentencing guidelines by a government commission were ruled to be a violation of ex post facto protections.
This is actually consistent with the history of the clause (It was understood to apply to criminal laws. Motions were made to change the wording to say it also applied to civil cases, they were turned down)
2. This is not a law, and may not even be administrative rulemaking (depending on what exactly they issued)
Of course, even if it was, you have a mechanism to challenge it if they hold you to it: the courts.
Doesn't that make this particular case a criminal one?
No, if the IRS doesn't think you've paid your taxes, they can attempt to collect the unpaid taxes.
OTOH, if the Department of Justice can prove in a criminal prosecution that you've violated the criminal provisions of the tax law, you can go to prison. If the IRS thinks you've done that (which is different from just not paying your taxes), they can refer you to the DoJ for prosecution.
This change to IRS guidance doesn't change the criminal provisions of the tax law that would apply were the government to prosecute you, therefore, the change is not a change to criminal law, retroactive or otherwise.
So you may end up with penalties and interest, but you have a high probability of avoiding prison absent doing something stupid like trading emails with your broker on how you plan to evade taxes.
I know a guy who didn't even file for five or six years. Eventually he got a lawyer and made a deal which had him paying some fraction of his original tax liability.
They'll put you in jail if you make them, but they don't really want to put you in jail. They just want your money.
Oh, this guy? After getting his sweet deal he didn't actually pay them. I think he may end up in jail.
Newly independent Americans in the late 1700s were very concerned with “this shameful abuse of power”, “those engines of oppression”, ex post facto laws. American businessmen were concerned with the macroeconomic effects and unfairness of interference with contracts by provincial/state legislatures.
One of the shameful abuses was the “paper money laws” that changed outstanding debts by permitting payment in paper money where the contract originally called for the use of gold or silver. There was widespread fear of the drying up of credit (in the context of an ongoing major depression at the time) as a result of concern that loans might not be paid back due to legislative intervention.
The dictionary of reference in 1739, Giles Jacob’s Law Dictionary, defined “ex post facto” as “a Term used in the Law, signifying some Thing done after another Thing that was committed before”.
Check out these sources: William Winslow Crosskey, “The True Meaning of the Constitutional Prohibition of Ex-Post-Facto Laws” (1947) 14 U Chicago L Rev 539
Roger W Weiss, “The Issue of Paper Money in the American Colonies, 1720-1774” (1970) 30 J of Economic History 770
Elmer E Smead, “The Rule Against Retroactive Legislation: A Basic Principle of Jurisprudence” (1935) 20 Minn L Rev 775
Harold J Krent, “The Puzzling Boundary Between Criminal and Civil Retroactive Lawmaking” (1996) 84 Georgia L J 2143
W David Slawson, “Constitutional and Legislative Considerations in Retroactive Lawmaking” (48) California L Rev 216
$25 billion is all employees money; Even if we exclude Simon's money ($14 billion) it is a ton. 35% avg annualized returns for 20 years. Simply stunning.
I don't think they meant every year was greater than 35%, I interpreted it as the overall average is more than 35%.
RIEF (Renaissance Institutional Equities Fund) RIDA (Renaissance Institutional Diversified Alpha) RIFF (Renaissance Institutional Futures Fund)
Of course there's probably still very large minimum investment thresholds.
OK, can someone explain this? Most people say that "you can't beat the market in the long term", "market is efficient", etc. So how can these people have done so well over more than 20 years??
Let me illustrate: We'll assume that trading returns have a binary distribution. Traders win or lose with equal probability. This is not a great model, but it's good for making ballpark estimates, because it overestimates the odds of a track record like Renaissances.
RenTec's Medallion fund has not had a down year in the past 25. The odds of this are at most 1 in 33 million, using our binary model. Survivorship bias does not begin to explain this; there have not been anything resembling 33 million hedge funds over the course of history. I think 30000 hedge funds is a fairly generous estimate.
Exercise: Suppose Medallion's returns are at least a standard deviation above zero, every year. Calculate the odds of this happening.
I've been investing for almost a decade, was lucky enough to sit out the worst in 2008, and I haven't had a down year in 10 years myself. This outcome was mostly luck on my part.
But in order for markets to be efficient there needs to be lots of constant arbitrage, and anyone who pulls off an arbitrage trade makes a profit. So maybe the Medallion fund just does a lot of the arbitrage that makes markets so efficient.
Without cheating that may be true.
My personal favorite is, "They trade on non-public information, and use a complex trade algorithm for the purposes of plausible deniability."
Another is, "They got lucky."
Another likely one is "The reporter did not calculate average annual return correctly."
One that I find dubious: "They actually are that good at investing."
For giggles: "They are using hedge funds to launder drug money."
I'm sure there are other ways to explain it, but that return just sounds way too good to be true.
I think they'll probably use a database that's "eventually consistent" ;-) I
The options also were attractive because they limited the risk of loss to the amount paid for each option, Renaissance said. “No other investment structure of which we are aware provides both high leverage and loss protection,” Renaissance said.
How can they have a 35% average annual return over the past 25 years and guarantee that you won't take a loss at all?
Reminds me of Madoff.
Well, their accountants and financial advisers might, since their work of structuring things to make sure most gains are long-term and in the right categories would be less valuable. But cutting taxes for all taxpayers -- including the rich -- on short-term games, and for most taxpayers on several categories of long-term games, may not that widely object to by people paying capital gains taxes in the first place.
All of these things are done to change the incentives about something, you can't expect them to be consistent without reference to those goals (and even then...)
"In a tax system with progressive marginal rates on annual income, treating income resulting from action over multiple years as earned in the year realized results in a artificial increase in the tax burden on those with that kind of income pattern. So, there's a sense in which treating capital gains earned over multiple years differently than capital gains earned over a single year or less (the latter taxed as general income), given the progressive nature of the tax system."
Of course, irregular income from other sources that isn't simple hourly work often also can result in windfalls from work over a longer period of time, or naturally intermittent so that an a large income one year is the result of a pattern of activity that also resulted in a smaller income other years. A simple way to address this that isn't specific to capital income is to tax all income equally, but allow, within certain bounds, income to be recognized for tax purposes, and associated taxes paid, in advance of realization and, perhaps in certain cases, deferred and recognized over a period of years after realization.
This would mitigate any overtaxing of cashing out events that took many years to "earn" the income, without favoring any particular kind of income or creating gameable thresholds.
Raise them prospectively based on the date of asset purchase, starting some specified time after the date the change is adopted. Problem solved.
They get the relative burden of taxation shifted off the of other sources of income, including the source they are using to get money to pay for retirement. (The specific effect depends on how that relative shift is used.)
In any case, if the current favorable treatment of capital is inequitable, there is no entitlement for people to expect it to continue going forward.
> They get the relative burden of taxation shifted off the of other sources of income, including the source they are using to get money to pay for retirement. (The specific effect depends on how that relative shift is used.)
You say that as though an increase in taxes here will be used to decrease taxes elsewhere. That does not match the reality of changes to tax law; tax revenue will simply increase, with no changes to taxes elsewhere.
> In any case, if the current favorable treatment of capital is inequitable, there is no entitlement for people to expect it to continue going forward.
There is no entitlement for government to apply arbitrary taxes and expect people to bear it, either, without providing a commensurate improvement in value.
This will happen regardless, as tax rates have nowhere to go but up in the USA (considering trillions in unfunded liabilities).
> And what, precisely, do they get in return for that added tax burden?
Civilization. That's exactly what taxes pay for.
Because when you've run up a giant pile of debt, the response should be "get and spend more money" rather than "spend less"?
> Civilization. That's exactly what taxes pay for.
As long as we're being snarky: product not as advertised, cost inflated well over initial agreement, return policy non-existent and no refund available. Would refuse to do business with again if not for geographic monopoly and coercion.
I don't get why more people don't do this!
My state, unlike most in the US and unlike the US federal government, at least has a requirement that all tax issues must be voted on by the general public rather than just passed by the legislature and signed by the governor.
The federal government, however, is the largest source of both taxes and onerous regulations, the most challenging and time-consuming to get involved with, and the most resistant to change.
(This is, oddly, a larger-scale version of the same problem Wikipedia has or many other organizations have: whoever has the most time to waste can be the most successful creating and thriving in a bureaucracy, winning by sheer volume rather than by merit. Or, alternatively: anyone who would want the job shouldn't have it, and anyone who should have the job wouldn't want it.)
As for "anyone who would want the job shouldn't have it," that's cute to some, but to me it's a tired thought-terminating cliche. It says "it's ok to be cynical, hur hur, and complain without doing anything about it," but does that get us any further?
I don't, however, think it ought to take a life-consuming effort to make it stop affecting a lot of people's lives. But unfortunately, it probably will. Not least of which because there are a large number of people who feel entitled to make it affect a lot more of people's lives.
In any case, though, I'm surprised that you say "I don't get why more people don't do this!"; I think it's fairly obvious why more people don't do this. If it were the kind of thing you could spend a few hours doing as a volunteer effort and have an impact, rather than the kind of thing you could spend your life on and possibly accomplish nothing, then more people might do it. On the other hand, if it were that way, then there wouldn't be as much of a problem to deal with in the first place.
Tax-sheltered vehicles such as IRAs and 401Ks already exist specifically for retirement. That's unrelated to the issue of preferential tax treatment for capital gains, which is not limited to retirees and is completely regressive, privileging those who already have access to capital over those who rely on their labor for income.
Regardless of the particulars of any taxation expectations, there's something about retroactive policy changes that seems to defy the spirit of law to me.
Renaissance pursued a strategy of tax avoidance and should be punished.