The David Swensen Asset Allocation

yale

The Yale Model

I have been looking for a good book about asset allocation, and I decided to pick up David Swensen’s book, “Unconventional Success.”

David Swensen should undoubtedly be regarded as a superinvestor. He took over Yale’s endowment in 1985 and achieved a return of 16.1% from 1985 through 2005. This is a considerable achievement, particularly considering the vast sums of capital that Swensen needs to deploy. He achieved these results by focusing extensively on moving away from the heavy allocation to bonds that the Ivy League had before the 1980s. His asset allocation is known as the “Yale Model,” which involves a widely diversified portfolio across asset classes and massive exposure to alternative asset classes, like private equity and venture capital.

“Unconventional Success” is not about how David Swensen achieved these returns for Yale. If it were, I wouldn’t have been as interested in the book considering I don’t run an endowment and never will. The book, instead, is aimed at small investors like myself.

The book is geared towards small investors who want to set up a simple, passive portfolio that will preserve their wealth and give a decent rate of return over the long run. Ben Graham would have called this “defensive investing” back in his day.

The Importance of Asset Allocation

Asset allocation is a critical topic. It is especially crucial for the FIRE crowd. While I admire the superhuman savings rates of FIRE bloggers, I think many of them are making a mistake by putting 100% of their savings in US equities, or an equivalent to VTSAX.

While US equities have been the best asset class over the long run, they are often prone to horrific drawdowns and “lost decades”. While over the long run everything will likely work out, 10 years of flat returns is a long time, especially when you are “retired” and don’t have an income.

Take the 1970s, a flat decade for US stocks. Stocks delivered a .26% real rate of return after inflation with a max drawdown of 40% during the ’73-’74 drawdown. The 2000s is another example. During the 2000s, stocks delivered a nominal return of 1.2% and a real return of -1.21%.

Again, everything worked out. If you had a long time horizon, you made out fine. If you were dollar cost averaging, it wasn’t a big deal. But if you’re retired, it is probably worth having a more diversified asset allocation.

It’s fully expected that elements of a balanced portfolio across asset classes will enter bubbles periodically (REITs in the 2000s, US equities in the 1990s, international developed in the 1980s via Japan), but in a balanced portfolio you won’t have 100% exposure to them. Another important aspect of a balanced portfolio is to have exposure to asset classes that will deliver a return in all economic environments.

US Stocks & US Bonds: The Classic Allocation

This is a blog about value investing and stock selection. That is not what Swensen’s book is about. In fact, Swensen advises strongly against individual investors picking their own stocks and believes that small investors don’t stand a chance. I do it anyway. As Han Solo says, never tell me the odds.

People often come to me for financial advice because they know I’m obsessed with this stuff. My advice is pretty simple: look at the Vanguard life strategy funds, look at the drawdown that happened in 2008, decide which one you could have lived with, keep piling money into it, and try not to look at it for 30 years.

Some mix of stocks and bonds are the standard asset allocation. Stocks supply the long run return, bonds deliver a boring rate of return but provide comfort to psychologically bear with the horrible 50% drawdowns that rear their ugly head every once in awhile. When it feels like all hope is lost, your bond allocation reaches out and says “come with me if you want to live.

Bonds give you the behavioral will to survive significant stock drawdowns. Stocks supply the long-term returns, with gut-wrenching volatility.

This has been the standard investing advice for decades. Swensen thinks that the small investor can do better. He starts by adding in asset classes beyond the standard US equity/bond allocations that dominate most individual portfolios.

The Core Asset Classes

Swensen describes his views on what he calls the “core asset classes” that deserve a place in a portfolio.

US Stocks: Swensen is talking about market cap weighted US equities, of course. No value tilts, trend following, etc. Good old market cap weighted US stocks. Since 1900, US stocks have delivered a 9.37% rate of return. This beats any other big asset class. Those high returns include horrific drawdowns, including a nearly 80% drawdown occurring in 1929-32.

US Treasury Bonds: Swensen is particular about bonds: he’s talking about US government bonds. He is not interested in corporate bonds, which he doesn’t believe deliver a high enough return relative to government bonds to justify their risk.

Bonds will not deliver a substantial rate of return but will protect investors during drawdowns. He also does not recommend owning foreign bonds, as they don’t have the creditworthiness of the United States.

He also recommends that half of an investor’s bond allocation go to TIPS or Treasury inflation-protected securities. TIPS were introduced in 1997 and are bonds that will increase in value if inflation rises. Conversely, they decrease in value if inflation falls. Swensen believes they deserve a place in the portfolio because they will protect against significant equity drawdowns, but they will also help the performance of the portfolio during a time of high inflation like the 1970s. Since 1997, TIPs have delivered a 4.78% rate of return.

Since 1997, we haven’t had a serious inflation problem, so this allocation hasn’t really had its chance to shine.

Meanwhile, the rest of the bond allocation exists primarily to shield the portfolio during drawdowns. The total bond market has delivered a 6.42% rate of return since 1900. They really help during drawdowns. From 1929-32, the total bond market provided a 24% return. During the global financial crisis, the bond market gained 11%.

Of course, that’s the total bond market. Long term US treasuries really shine in drawdowns. In 2008, while every other asset class felt like Britney Spears circa 2007, Vanguard’s long term treasury fund (VUSTX) gained 22%. This occurs because, during a meltdown, investors flock to buy US treasury bonds for safety and the Fed is typically pushing up bond prices to drive down interest rates.

International Developed Equities: Swensen recommends diversifying outside of the United States in developed international equities. Developed meaning that they are no longer growing rapidly and are in a similar position of economic development to the United States. Think of Western Europe & Japan. Since 1976, this asset class delivered a 9.04% rate of return. Like the US market, they are subject to horrific drawdowns.

Emerging Markets Equities: Emerging markets are economies that have recently entered a level of industrial level of development from an agricultural subsistence economy. The ultimate example in recent years is China. Since 1976, emerging markets have delivered a 9.04% rate of return.

Like international developed, they are not entirely correlated with US equities. From 2000 to 2010, a period of time when US equities lost money, they delivered a 10.61% rate of return. Since 2008, they have only achieved a 1.56% rate of return. For this reason, they fit nicely into a portfolio. Historically, they tend to offer high returns when US equities are not performing well.

Much of this has to do with valuations at the start of the decade. US equities were merely much cheaper than emerging markets or developed markets in 2010. Conversely, US equities were ridiculously more expensive than the rest of the world in 2000, which is why they sucked for the decade of the 2000s. The nice thing is that all of the world economies don’t usually enter bubble territory at once.

REITs: REITs, or real estate investment trusts, are companies that own a broad portfolio of income-producing real estate assets. Since 1970, REITs have delivered a 10.99% rate of return. This is another asset class not wholly correlated with US equities that can provide a decent return. Real estate also tends to increase with inflation, which is why Swensen believes REITs deserve a place in a portfolio for their inflation protection.

In particular, they held up well during the tech meltdown. From 2000 through 2002, REITs delivered a 13.78% rate of return while US equities were cut in half. Their worst moment was during the housing crisis when they declined as an asset class by 47%.

REITs also shined during the 1970s inflation, which led to an abysmal return for US equity investors. During the 1970s, REITs delivered an 11.32% rate of return.

swensen

The Building Blocks

According to Swensen, the “core asset classes” are the critical building blocks of a portfolio. They will rarely all work at once. An investor will be diversified across asset classes, and the different asset classes will all work their magic at different times.

The bonds exist as protection during stock drawdowns. TIPs have the added benefit of protecting an investor during inflationary periods. Longer termed treasuries will protect the investor during panics or truly horrific deflationary episodes, like 1929-32.

Meanwhile, the rest of the portfolio is divided up into asset classes that should deliver a very high rate of return. The nice thing about the asset classes that Swensen selected is that they don’t all perform well at the same time.

When stocks absolutely sucked during the 2000s and barely beat inflation, REITs and international equities still performed very well. During the 2010s, US stocks did great while international stocks sucked. If we have a repeat of the 1970s, stocks will probably be crushed, but REITs and TIPs will pick up the slack. Ideally, the approach should help an investor avoid “lost decades” like US stocks not delivering any return in the 2000s. You’ll always hate something in your portfolio, but you’ll also probably have something to love in your portfolio.

Most importantly, you can buy all of these asset classes directly through low-cost mutual fund and ETF vehicles. You’re not doing anything fancy with options or hedging. You don’t need to pay excessive fees to someone to do this for you. All of these asset classes are available in most 401(k) lineups, as well.

An added advantage is that these can all be easily managed. A small investor can do this themselves without any help.

DIY Asset Allocation

Swensen recommends that an investor hold no more than 30% of an asset class in their portfolio. This allows the diversification to work. It prevents one asset class from dominating the portfolio. Swensen’s specific recommendations are as follows:

US Stocks – 30%

Foreign Developed Stocks – 15%

Emerging Markets Stocks – 5%

REITs – 20%

US Treasury Bonds – 15%

US Treasury Inflation Protection Securities – 15%

Here is how you can construct this portfolio using simple, low-cost Vanguard ETF’s. You could also do this with mutual funds. My preference is for ETF’s because they maximize your tax benefit. The simplicity is key here: you could rebalance this once a year in your underwear. Investing is one area of life where it actually pays to be a little lazy and not meddle too much.

US Stocks – The Vanguard ETF for US stocks is VTI.  The mutual fund equivalent is VTSAX. Many of the FIRE bloggers recommend putting 100% of your net worth into one of these. Personally, I think that’s fine if you have a long time horizon. However, if you’re already retired, you probably don’t want to be in an asset class that could potentially do nothing for 10+ years and would benefit from diversifying into more asset classes. The expense ratio is a dirt cheap .04%.

International Developed – The Vanguard international developed ETF is VEA. The expense ratio is .07%.

Emerging Markets – The Vanguard emerging markets ETF is VWO. The expense ratio is .14%.

Side note: Alternatively, you could replace the three equity ETF’s above with one ETF – Vanguard Total World Stock Index, VT. This is a market cap weighted index of all stocks in the entire world for an expense ratio of .1%. Currently, the portfolio is 60% in US stocks, 10% emerging markets, and 30% international developed. If this ETF is 50% of your portfolio – you will get right to Swensen’s recommended allocations. 30% of your portfolio will be in US stocks, 15% will be in international developed, and 5% will be in emerging markets. Of course, these allocations won’t remain static and will fluctuate based on the performance of countries within the global portfolio, but it seems like a much easier way to simplify a Swensen portfolio with one equity ETF instead of three.

REITs – The Vanguard real estate ETF is VNQ. The expense ratio is .12%.

TIPs – The Vanguard TIPs fund is VTIP. The expense ratio is .06%.

Treasuries – Swensen doesn’t specifically recommend long-duration treasuries in the book. My preference for long-term treasuries is merely because they will do the best when the markets fall apart, which is the whole reason they’re in a portfolio. Vanguard doesn’t have an ETF for long duration treasuries, but they do have a mutual fund, VUSTX. The expense ratio is .2% on the investor shares. iShares has an ETF for long duration Treasuries, TLT. The expense ratio on that is .15%. If you want to follow Swensen’s advice and own treasuries of all durations, there is also an iShares product called GOVT.

Backtesting Swensen

How have Swensen’s allocations performed over time?

Since 1985, when the data is available for his recommended mix of asset classes (the TIPs returns are made up from 1985 through 1997 based on inflation rates), the Swensen portfolio has returned 9.77%. After inflation, 6.99%. $10,000 invested in the Swensen allocation in 1985 would be worth $238,007.

The worst year for the portfolio was 2008, in which it lost 26.58%. The total US stock market lost 37% during that period, and REITs were cut in half, while a 60/40 portfolio lost 20%. Before the global financial crisis, the worst year for the Swensen portfolio was a loss of only 6.92%. This occurred during the early 2000s meltdown when US-only equity investors saw their portfolio cut in half.

The portfolio really shined during the 2000s, when US stocks did nothing. The portfolio achieved a return of 6.88% from 2000 to 2010, or 4.34% after inflation.

TIPs didn’t exist in the 1970s, but substituting Swensen’s bond allocation for intermediate-term treasuries, the Swensen portfolio would have delivered an 8.26% rate of return and a small return of .41% over inflation. This is during a period where double-digit inflation led to losses across nearly every asset class. Moreover, the Swensen portfolio did this when bonds were crushed due to rising interest rates. The fact that it kept up with inflation at all is impressive. If TIPs existed back then, I would guess that the portfolio would have done even better.

Since the book was published in 2005, the portfolio has continued to do well. It has returned 6.68% or 4.56% after inflation.

Other Considerations

Taxes – Avoiding taxes is a significant consideration for most. The bond and REIT allocations will generate most of their return via dividends, not capital appreciation. If you hold them in a taxable account, you will have to pay taxes on the payments every year. To avoid this, you might want to consider carrying them in a non-taxable account like an IRA or 401(k).

Other asset allocations – Swensen’s allocations aren’t the holy grail of passive investing. There are other alternatives. Meb Faber and Eric Richardson cover a lot of interesting allocations in The Ivy Portfolio. Meb even has an ETF that covers his recommended asset allocation, which he dubs the global asset allocation, GAA. Another great book on the subject is DIY Financial Advisor from the great people at Alpha Architect.

Tweaking it – If you’re like me, you like to test and tweak things yourself. For instance, my preference is to gear my passive equity allocations towards value. US small cap value is the best performing asset class of all time. Since 1927, small-cap value has delivered a 10.93% rate of return. Why not gear more of my portfolio towards that? Or, you might want to set aside some of your portfolio to buy a portfolio of value stocks, like I did. If that’s not your bag, maybe you want to try trend following and momentum, even though I think that’s hocus pocus and nonsense.

There are some great resources available to experiment with your own allocation. One great resource is Portfolio Charts, a helpful visual resource where you can analyze different asset allocations from the Vanguard Three-Fund Portfolio to the classic 60-40, and it includes an analysis of the Swensen allocation. The author also maintains an excellent blog discussing asset allocation.

If you want to get deeper than that and experiment with more asset allocations, there is this resource at the Bogleheads forum. This is what I used to generate the return data listed through the blog post. You can experiment in that Excel spreadsheet with all different kinds of allocations.

Conclusion

There is no holy grail to investing. If you’re looking for a simple, low cost, easy to implement portfolio that doesn’t require a lot of work – the Swensen portfolio is a great place to go. Odds are, you’ll avoid “lost decades” and will preserve your capital over a long period.

As I stated before, 100% US equities might work for the FIRE bloggers, but I really think more of them should consider a more diversified asset allocation. They can’t afford lost decades, after all.

If you’re crazy like me, you also like picking your own stocks or pursuing a pure quant based approach like the Magic Formula or Acquier’s Multiple. Personally, I explicitly set aside this account that I track on the blog to buy and sell shares. My 401(k) and taxable account are in more of a diversified asset allocation like Swensen’s. My equity allocations are (obviously) geared towards value.

The bottom line is that you have to find what works for you. Good luck on your journey.

Random

This wouldn’t be one of my blog posts if I didn’t throw in something completely random. This is pretty funny.

PLEASE NOTE: The information provided on this site is not financial advice and it is for informational and discussion purposes only. Do your own homework. Full disclosure: my current holdings.  Read the full disclaimer.