Asset Allocation Conundrum

When I decided to focus on a diversified ETF strategy, I hoped asset allocation questions were settled science.

Nope. Not even close.

Accepting Lower Returns

One aspect of this that's a hard pill to swallow is that it will likely force me to accept lower returns. When I buy an individual stock, my hope is that it will appreciate by 15% a year for the foreseeable future. Naturally, volatility won't make that a smooth ride, but generally 15% is the goal.

Expecting a 15% CAGR when you're allocating to a basket of assets is foolish. After backtesting various portfolio types, it's probably wise to expect a 4-9% CAGR. Some years will be better, but due to overvaluation and volatility, sometimes there's no escaping bad returns. There might be an extended period of 0 return.

If I can get 15% on a stock but only 4-9% on ETFs, then why get the ETFs? Well, it's because the stocks aren't guaranteed to work. Additionally, they expose me much more to my own thinking errors, behavioral mistakes, and biases. Perhaps my analysis is simply wrong or under-baked. Basically, I need some money set aside into broad buckets that will perform well enough in case my stock picking doesn't work out.

That's where asset allocation comes in, and - even accepting lower returns going forward - it's tricky.

Why Not Just Do Something Lazy?

So what would be the, you know, "Ah, screw it" portfolio?

As a baseline, there's J.L. Collins 100% allocation to VTSAX (ETF: VTI). Lazy and easy, and you won't feel a lot of FOMO (fear of missing out) since it's the whole U.S. stock market. It's also very hard to beat:

But what if you want some "hold onto your butts" assets for the scary times? Something like the Bogleheads' 3 fund portfolio fits:

For sure, lazy has a lot going for it. It would be easier for me to manage, and should I get hit by a car, it will be easier for my wife to manage. There are some portfolio varieties that just have too many funds in too many strange percentages. If it requires a computer to manage, then it's probably not the best choice.

And the lazy portfolios don't perform strangely. If you spend years envying the S&P 500, how long can you reasonably hold out before you just buy the S&P 500?

That said, are there ways to achieve slightly better risk-adjusted returns without it becoming too complicated?

International

I should have been prepared for difficult questions since I've been an active listener of Meb Faber's podcast. He has idiosyncratic views about asset allocation. For example, he argues effectively that a global allocation is not only valuable but even dangerously underutilized in many modern portfolios.

To make a long story short, being concentrated in one country in a market-cap weighted portfolio leaves you open to major country risk as well as valuation risk. There will be periods of underperformance, and there's the risk of a total country disruption that leads to a total loss.

What's hard to get over, however, is that international additions to a pure 100% US market allocation have been a performance drag in recent memory. It's one thing to know that allocating all your assets to a single country (even the US) can be risky, and it's another to actually allocate money to underperforming geographies:

Adding international exposure looks like a leap of faith based on the following ideas:

• The US is likely overvalued relative to international markets, which may lead to sustained international outperformance during times when the U.S. is working through overvaluation.
• The existential risk of single country concentration is high enough that putting up with potential lower returns is worth the risk. Countries don't stay on top forever, and whole country's stock markets have gone to 0.
• Adding international adds even more diversification.

With those ideas in mind, I will probably add an international component.

Drawdown Protection Portfolios: 60/40, Weird, and Permanent

If I want lower drawdowns in a lazy way, there's not much lazier than the 60/40 or 40/60 portfolio:

One of the accounts I follow on Twitter is ValueStockGeek. While occasionally, he'll put out some information on a specific company he's interested in, the most surprising content he writes is about his Weird Portfolio. I encourage anyone interested in this stuff to read what he's written on it, but long story short it's:

• 20% US Small Cap Value
• 20% International Small Cap
• 20% Gold
• 20% REITs (divided between US and ex-US)
• 20% Long term treasuries

It's his variation on Harry Browne's Permanent Portfolio, and it's more aggressive than Browne's risk-averse allocation (25% US stocks, 25% Long Term Treasuries, 25% Gold, 25% cash).

Notice the slow and steady return of the blue line (the Permanent Portfolio) vs the more jagged yellow line (the S&P 500), with the red line (the Weird Portfolio) somewhere in the middle.

Both alternative portfolios are trying to consider inflationary and deflationary environments, and how those periods impact the various components. When both portfolios succeed, they leads to lower but steadier growth with much gentler drawdowns. The Sharpe ratios are considerably higher than a pure stock allocation. The Permanent Portfolio's returns are low but with much lower risk, while the Weird Portfolio has more acceptable total returns.

Backtests point to something like an 8-9% return for the Weird Portfolio. In the Great Recession, the drawdown was higher than the Permanent Portfolio's, but it was much lower than a 100% stock allocation. Combine that with a 9% return, and it feels like a revelation.

My concerns with it however are:

• Small cap value outperforming over time is necessary for the growth to be satisfactory. It hasn't out-performed during this last decade, though its longer term record is very good:

• The drawdown protection via gold and treasuries has to actually work.

Nevertheless, I find it compelling despite my concerns and will integrate some of this into my own approach.

Other Compelling Portfolios and Final Thoughts

Look around enough, and you'll see all sorts of smart portfolio constructions. Take a look at the Ginger Ale Portfolio for one idea. For my taste, it's too many ETFs, but to each their own. Or stroll through the Boglehead forums to read intelligent debates about portfolio construction.

Nothing is guaranteed, and we have to make best guesses about our own psychology and how best to navigate an unknowable future based on the available research and how assets have behaved in the past. Avoiding blunders is paramount.

The strongest takeaways for me are:

• U.S. Market exposure is basically good enough on its own. It prevents FOMO and will probably perform well. It has risks though.
• Some "hold onto your butts" assets make sense.
• A tilt towards small cap and value makes sense.
• Too many assets gets unwieldy.
• Some international exposure is likely worth it despite recent underperformance.

This only begins to scratch at the surface of asset allocation decisions that someone could obsess over. I think I have a basic plan, when I make a decision I'll write more.