Tail Risk – A Quick Guide

I’ve been incorporating ‘Tail Risk’ ETFs into my portfolios for some time now, specifically TAIL in the Model Portfolio and TAIL, CAOS, and BTAL in my personal portfolio. Given the current market conditions, it feels like an opportune moment—or perhaps the worst time, if you haven’t already invested—to provide a brief recap. If you’re not yet invested, it might be too late for this cycle.

Tail Risk Protection

Jason Buck says that Tail Risk protection strategies are characterized by 3 elements, out of which you can only pick 2: convexity, bleed and surety.

• Convexity: how much the strategy returns when stocks are in a severe drawdown. The higher the convexity, the lower we have to allocate to the strategy because it creates more return for a unit of capital.
• Bleed: how much the strategy loses when markets are calm.
• Surety: will the strategy deliver when markets tank?

The TAIL ETF, for example, has convexity and surety but “pays” for them with a high bleed.

CAOS is the Alpha Architect Tail Risk ETF; in reality, AA is just the white label ETF provider, the engine of the product is managed by Arin Risk Advisors. CAOS has high convexity and low bleed, but the price for low bleed is not 100% surety. Think about it as market timing: the fund doesn’t offer the same level of protection all the time, it buys insurance only when the price is right (in reality, it is a mix of buying and selling insurance).

BTAL is the AGF US Market Neutral Anti Beta Fund. BTAL’s objective is to provide a consistent negative beta exposure to the U.S. equity market. BTAL strives to achieve this objective by investing primarily in long positions in low-beta U.S. equities and short positions in high-beta U.S. equities on a dollar-neutral basis, within sectors. BTAL has high surety and low bleed, but no convexity.

How to manage Tail Risk

Just like any other insurance, managing tail risk protection can be challenging. Paying premiums (the bleed) when nothing happens simply diminishes your portfolio returns. While attempting to time the market is risky—you may end up without the protection precisely when you need it most.

Let’s delve into the mechanics of these ETFs and explore how we can effectively utilize them.

CAOS

I start with CAOS because it is the one where there’s very little to say.

The above picture is taken from this Alpha Architect presentation. The strategy behind CAOS was previously run as a mutual fund with ticker AVOLX; unfortunately, PortfolioVisualizer has removed AVOLX from its database and TestFol.io never had it. So, no backtest.

For what is worth, this is how CAOS performed since its launch. The advantage compared to a “simply reduce your risky assets exposure”, aka invest in cash, is pretty clear. CAOS is cash WITH insurance. Probably too good to fully trust but worth at least a small allocation. AVOLX demonstrated the same characteristics with an additional 10-year live performance: this is definitely comforting.

If you want to have your cake and eat it too, the strategy is to borrow and invest in tail risk protection. If you manage to keep the borrowing cost close to the risk-free rate, you can basically get full returns from your risky assets (you are not sacrificing space in your portfolio, the leverage is allowing you to allocate to tail risk protection) and have some insurance when shit hits the fan.

CAOS’s performance so far, when compared to CASHX, should conveniently show how this works in practice. The beauty of this strategy, if it continues to perform in the future as it did in the past, is that you do not have to time anything, just rebalance from time to time as you would normally do with your plain-vanilla portfolio.

BTAL

What I just wrote is probably easier to understand in the following backtest using BTAL.

I AM NOT ADVISING YOU TO BORROW 20% AND ALLOCATE IT ALL TO BTAL! I just used that high % in order to make the results ‘visible’. You can have a benefit with a lower % (obviously).

Adding BTAL reduces volatility and improves the portfolio Sharpe ratio. Even with a cost of borrowing of 1% above the cash rate. And that’s it: no need to time the market or any other adjustment. Simply add BTAL and let it cook.

If you are thinking that the improvement is little…I would rebate: it is still ‘free risk management’. What happens if the toy breaks up and stops working? This is a more relevant question. Am I collecting (risk-adjusted) pennies in front of a steamroller? The short part of the strategy can always blow up for reasons independent from the strategy itself: shorting stuff is more complicated than buying xyz – you need a willing owner to lend you the stuff to short at a reasonable price. Around June 2020, so after the stock market crash, BTAL dropped for a few days just to recover immediately after. There is always this risk but I think it is worth taking, especially if you have a sub-10% allocation to the fund.

TAIL

With TAIL, I tried to mimic a different strategy, closer to “market timing”, for a few reasons. TAIL is a first responder type of protection: it works when stocks crash FAST. It definitely does not work in a 2022-type of decline; but for these scenarios, I have other responders in my portfolio, i.e. trend following funds.

The strategy described in the TestFol.io backtest rebalances when the portfolio elements diverge by 30% or more from the starting allocation. I used this to mimic what I am doing in reality – only I can model it more softly.

So yeah—when the market’s down 20%? I’m already thinking of ringing the register. Down 30%? I’m trying to cash out the insurance, not rebalance it. In theory, I’d love to be flat TAIL by the time we bottom. In practice? You’ll always be a little late, or a little early. I’ll take early.

Here rebalancing is a razor’s edge. Do it too often—death by paper cuts. Wait too long—you miss the payout. There’s a narrow Goldilocks zone. Most people rebalance because they like doing something. This isn’t that. This is about unloading when the bid for insurance is highest. Selling panic.

The strategy also reduces the Sharpe ratio differential when I increase the leverage cost to 50bps or more.

There are only a handful of environments where TAIL actually helps. Most of the time? Value bleed. But…

It’s insurance for the other insurance—CAOS, BTAL. It’s the third wheel, and not the fun one. Statistically the worst performer. Still—it’s the one you want when things go really sideways. Like, “phones down, what’s my NAV” sideways.

If there’s ever a margin call—I don’t use much leverage, but still—TAIL is the one piece that’s probably up. Not “winning,” but less bad. That matters. Having something green when everything else is bleeding buys you time. Buys you clarity. And time and clarity are expensive when you’re losing.

Most of the time, this thing looks like dead weight.

But when you need it?

You really need it.

Now the part that most likely will make your head explode

The above considerations work for the generic investor out there. In Switzerland, I can deduct interest expenses from my income. This means I can buy this type of insurance for free and push the leverage a bit more; or, if you want to see it from another angle, I can match your unlevered portfolio return with a lower volatility concoction.

What I am reading now:

Follow me on Bluesky @nprotasoni.bsky.social