Forums  > General  > Short-Term Cap Gains Taxes and Compounding  
Page 1 of 1
Display using:  


Total Posts: 348
Joined: Jan 2015
Posted: 2017-01-17 06:22
(Pre-emptive apologies for the Amero-centric thread. International discussion obviously welcome, but my perspective's American.)

Short-term trading is obviously tax-disadvantaged because it's taxed as ordinary income, which roughly works out to ~40% vs ~20%. But the more subtle disadvantage is that taxes are harvested every year. For high return portfolios this significantly impairs the gains from compounding. Over long periods, the compound drag can actually be substantially worse than the higher rates.

A toy example: Let's say you're running a short-term trading strategy with 50% annual ROI. After a 40% tax rate this nets out to 30% a year. Compounded over 15 years, a $1 million initial investment would grow to $51 million. Now let's say you had a tax-shield which keeps the same rate, but defers the tax bill until the end of the period. The $1mn compounds to $437mn pre-tax in fifteen years. After paying 40% of the realized profits, you'd net out $262 million. More than 5 times the final capital of the base case.

Some imperfect ideas to mitigate this, and their major limitations. (Not a tax lawyer, so obviously any of the below certainly doesn't constitute *advice* and may be (very) incorrect):

- Charitable Remainder Trusts: This seems like the best option. Allows the investor to defer taxes until trust distribution. Biggest disadvantage here is having to preset a fixed distribution schedule, which substantially limits flexibility and liquidity. Also you lose all the money if you die before the trust expires (meaning you need to keep upping life insurance coverage as the trust grows).
- IRAs: This seemed to work for Romney, Thiel and Dustin Moskowitz who all compounded $5,000 contributions into $100+ million portfolios. But there's pretty substantial limitations for traders: no margin, no short-sale, no trading more frequently than every three days.
- Puerto Rico Act 22: US citizens who spend 180 days in Puerto Rico are totally exempt from cap gains on trading. Obviously this not only limits the compounding drag, but any tax liability. (In the toy example, you'd wind up with the entire $437mn). Your wife and kids may be slightly less than enthused.
- Reinsurance Reserves: Convert the strategy capital into reserves at a reinsurer. Capital gains is timed based on when the reserves are sold, not when the underlying securities making up the reserves are sold. Don't know how viable this is unless you're already a huge investor like Paulson. Also not sure how extensive the risk limits are for insurance reserves.
- Basket options on the portfolio: I know RennTech was doing this for a while, but my understanding is the IRS has pretty much shut this down.
- Find a tax-exempt entity to run the strategy. Donate the original capital, have your management company run the strategy, and retain a high incentive fee. Even if you end up paying out 50% of the final capital, you're still way ahead of the base case. If you stop trading earlier rather than later, you're out the entire original principal though.

What else am I missing here? I'd figure a quant community would have a lot of experience here. Mainly since they tend to mostly engage in short-term trading, and have high return portfolios (and hence high tax-compounding drag).

Good questions outrank easy answers. -Paul Samuelson


Total Posts: 458
Joined: Jul 2008
Posted: 2017-01-17 10:06
Or maybe pay the damn tax and stop being a free-rider?


Total Posts: 348
Joined: Jan 2015
Posted: 2017-01-17 10:23
Keeping the same rate, but deferring the realization actually means that the US Treasury comes out ahead. Consider the original toy example. Assume the government's long-term discount rate is 3% (current 30 year yield). In the base case (taxes paid every year), the IRS collects $33 million over the 15 year life-time. Discounting the cash flows yields an NPV of $23 million in tax liability.

In contrast in the deferred scenario the IRS is collecting a single tax bill at the end of the life time. But it's substantially larger than the base case: $174 million in 15 years. That's an NPV of $111 million in tax liability.

The way the math works, Uncle Sam actually comes out ahead in deferring tax realization for any investor with an ROI higher than [discount rate] / [capital gains tax rate]. For 3% treasury yields and 40% short-term cap gains rate, that's a 7.5% hurdle rate. If you can consistently out-perform 7.5% annualized returns, it's actually patriotic to defer tax realization.

Good questions outrank easy answers. -Paul Samuelson


Total Posts: 458
Joined: Jul 2008
Posted: 2017-01-17 10:35
"In contrast in the deferred scenario the IRS is collecting a single tax bill at the end of the life time"

What is the lifetime? Is it capped?


Total Posts: 348
Joined: Jan 2015
Posted: 2017-01-17 10:47
Sorry. To be clear, I was referencing the toy example from my original post. In this case, the arbitrarily assumed lifetime of the strategy was 15 years. After which presumably the fabled investor retires from active management, cashes out into a low-fee Vanguard fund, moves to Boca (in the low money scenario) or Palm Beach (in the high money scenario), and takes up golf and watching NCIS re-runs.

Obviously this is just a make-believe scenario. And as far as I know, no exact tax structure with the described properties exist. (Plus can the strategy actually continue to compound without hitting capacity constraints? What about generalized performance decay, drawdown risk, etc.? What if the investor dies before realizing, stepping up the cost basis through inheritance?). But the basic principles are all the same. Tax deferment can actually be a net benefit for the IRS, if the investor re-invests and generates high return.

Good questions outrank easy answers. -Paul Samuelson


Total Posts: 1016
Joined: May 2004
Posted: 2017-01-17 11:44
Well, Puerto Rico is not looking too bad. Maybe the weather is too cold during the winter where you live and you need to spend half of the year in a warmer climate.

Otherwise, give up your US passport and move to Dubai, Singapore, etc... where your effective tax rate can be pretty close to 0.

"Earth: some bacteria and basic life forms, no sign of intelligent life" (Message from a type III civilization probe sent to the solar system circa 2016)


Total Posts: 465
Joined: Apr 2005
Posted: 2017-01-17 11:44
There are two companies trying to push the reinsurance reserves option - and pushing it as a platform - so more pay as you go (some up front $ costs - but a fraction of costs of doing it yourself, two sigma/paulson/greenlight style).

Drop me an email off line if you want more details - we've been looking at it (although not quite there)

Nuclear Energy Trader

Total Posts: 1284
Joined: May 2004
Posted: 2017-01-17 20:52
Regarding IRAs ... That bit about no trading more often than every 3 days?
Is that really true?

flaneur/boulevardier/remittance man/energy trader

Founding Member

Total Posts: 5068
Joined: Mar 2004
Posted: 2017-01-18 01:17
Welcome to the shit my friend. IMO, we should look at spending and wealth/consumption taxation before any more wage taxation. Most people at 40% are effectively paying more when you factor in AMT, etc. and all the other taxes and post-tax expenses.

Look into life insurance. A number of people I work with use life to invest and get tax deferred accumulation that they intend to withdraw from later.

In general, anything you can do to reduce your W-2 wages is key. Defer as much as possible.

Of course what I just mentioned really only is material for circa 500k-1m. After maybe $1m of W-2 the pain starts up again (depending on what options you have for deferral) and you need to get more creative.

Nonius is Satoshi Nakamoto. 物の哀れ


Total Posts: 74
Joined: Nov 2010
Posted: 2017-01-19 13:44

Yes, it is. You have to wait for the funds to settle before you can trade them again, which is T+3. Like he said, no margin.


For the tax-exempt entity, it can't be a simple charitable foundation. They also have limitations on margining. I don't remember what they are, but I remember it being a deal breaker.


Total Posts: 91
Joined: Sep 2015
Posted: 2017-01-21 22:04
Why not just move to PR and enjoy life?

I have been planning this for myself in the coming years.

I think a separate interesting question is at what AUM is this actionable/sustainable at your toy 50% ROI.


Total Posts: 348
Joined: Jan 2015
Posted: 2018-02-19 10:07

1) Anyone aware of new opportunities from Trump's plan?

From what I've seen, the lower corporate rate might offer a petit opportunity. When corporate rates were 35%, it almost always made sense to structure investments as a passthrough. But at 21%, you might get more juice out of compounding than you pay in double taxation.

In the toy example from my original post, structuring the strategy as a non-passthrough corporate would produce a post-(corporate-)tax annual ROI of 39.5%. If the initial $1 million grew inside the corporate structure for fifteen years, it would hold $147 million. Assuming dividends are taxed at a personal rate of 23%, you'd get paid out $113 million. More than double the straight pass-through option.

The passthrough deductions don't seem like anything great for traders. Even if you fully utilize the 20% deduction, it lowers the top marginal rate from 40% to 32%. And getting that deduction is no easy feat. You either have to 50X the deduction in real capital expenditures, or 4X the deduction in payroll. Obviously neither of those really apply to a trading operation. Best I can think of is finding some corporate with a lot of people on payroll, who's not utilizing the deduction. Then cut some deal to "employ" all their employees as a sort of middleman "service provider".

Finally there's the 100% CapEx deduction. Maybe there's some interesting games to play here. What I can think of is deferring trading-related taxes with offsetting CapEx. E.g. if you make $1 million in 2018 from trading, spend $1 million on some long-lived asset with steady cash flow. Like a commercial building or something, which you can lever up. That sets your tax liability to zero for the year. Assume you can borrow 90% against the asset, you're only taking $100 thousand out of the trading operation. It defers most of the trading related taxes, so you avoid most of the tax-compounding drag.

Anything else interesting in this front?

2) Related to plain old pre-Trump tax law... Anyone have any color on using the Active Financing Exception to Subpart F? Normally a majority-American owned offshore company is "collapsed" into a passthrough entity for tax treatment purposes. But as I recently learned the law requiring this, Subpart F, has an exception related to "active financing".

The classic example is a bank engaged in genuinely offshore activities. But at the very least it seems like a securities brokerage is also covered (as long as it only touches non-American securities). Assume you're running a quant trading operation that covers international markets. I would think you segment off the non-domestic trading into an offshore company, and use this to shield income under the active financing exception. Particularly if you're closer to the HFT end of the spectrum, where the operations do seem quite similar to what a securities brokerage does. However I couldn't find any specific guidance on this topic, so curious if anyone has looked into this?

Good questions outrank easy answers. -Paul Samuelson


Total Posts: 1016
Joined: May 2004
Posted: 2018-02-20 11:33
Not an expert in US matters, but the IRA route is interesting. Could you license some software to a e.g. Dubai-based company that does the trading? The money compounds tax free inside the Dubai company, and if you hold it in your IRA I suppose your exit would be tax free too.

I suppose it can't be that easy, but then you see what Romney did...

"Earth: some bacteria and basic life forms, no sign of intelligent life" (Message from a type III civilization probe sent to the solar system circa 2016)


Total Posts: 348
Joined: Jan 2015
Posted: 2018-02-27 19:13

That's a great idea, and you piqued my curiosity. The good news is that it seems like Roth IRAs are exempt from Subpart F. So you setup an offshore C-corp in a jurisdiction without corporate taxes, and avoid both US income and corporate taxes. This way you also get around the leverage and settlement restrictions. As long as the corporate entity is limited liability, you're not pledging IRA assets as collateral.

The hitch comes from avoiding "prohibited transaction". Basically the IRA cannot engage in any transactions with the IRA holder, immediate family or their fiduciaries. The logic here is to prevent someone from transferring value into/out of the IRA by doing something like selling an expensive asset to the IRA for $1 or having the IRA pay themselves a salary.

So you definitely cannot receive management or incentive fees on the investments. And even leasing the software from yourself or an associated entity also doesn't smell kosher. However, I think there may be a couple of workarounds. One is to "release" the software under the MIT license or some other copyleft. Just don't publish it anywhere. In this way, there's no actual transaction. The IRA entity is simply utilizing an open-source piece of software, that just so happened to be created by the disqualified party.

Another would be to do something akin to Romney. He put his interest in the partnership inside the IRA. Since there was no actual revenue streams associated with the partnership at the time of formation, the tax rules allowed him to value his interest at $1, well under the contribution limit. I think you could also do something with the software license itself. As long as it's not being actively used at the time. The downside is that it limits you from using the software in any context outside the IRA, because that would be a prohibited transaction.

However I should say once you get this far out, with this or any of the other more out there proposals, there isn't really so much "law" in the classical sense. The statutes are ambiguous, case precedence is sparse, and the underlying concepts don't really map on to these scenarios. (It's not even really clear if capital gains from quant trading qualifies as "active" or "passive" incomes.) You can run through the wording of the law umpteen times with tax accountants and lawyers, but it's going to basically come down to the subjective opinion of the tax court judge.

Good questions outrank easy answers. -Paul Samuelson


Total Posts: 1
Joined: Mar 2018
Posted: 2018-03-01 06:15
Be it at college or school level, the tutors are knowledgeable with years of experience and the admirable degree to ascertain that your Get the Dissertation help is in good hands. They are accessible to answer your questions 24/7.
Previous Thread :: Next Thread 
Page 1 of 1