This was written in 2015 and no longer reflects all my current thinking, though I think still makes some good points.

Many in our community are investing money to donate later, as well as saving for retirement and emergencies. Here’s some mistakes I’m concerned they’re making when investing.

I’m not a qualified financial advisor, and this should not be taken as investment advice. For speed, I’m also not referencing all my claims and this piece isn’t as thoroughly researched as normal – I just want to get the ideas out there. Please do your own research before making any investments. This post is based on personal interests of mine, and was not written in work time.

This post is aimed at people who already understand the basics of personal finance and investing. Some starting points for an introduction are here and here.

In summary:

1. Don’t expect to earn 7-10% returns from US equities. It’s more likely to be 1-7%. Adjust your assumptions about retirement savings and giving now vs. giving later calculations accordingly.
2. The baseline portfolio is the global market portfolio, roughly 40% international stocks (half US, a quarter emerging and a quarter other developed markets), 20% corporate bonds, 30% international government bonds, and 15% real assets. If you don’t think you can beat the market, this is much closer to what you should invest in than 100% US equities.
3. Divide your savings into a personal component and an altruistic component. Make sure you’re saving enough in the personal component to cover emergencies and your retirement. 10-20% of income is a rough rule of thumb.
4. Don’t take too much risk in the personal component. Allocate between the global market portfolio and cash depending on your risk tolerance and time horizon. e.g. If you have a time horizon of ten years and a low risk tolerance (~5% volatility is OK), then hold about 50% in cash and 50% in the global market portfolio.
5. With the altruistic component, depending on which causes you support, you should probably take substantial risk, perhaps going all the way to a leveraged portfolio.
6. Advanced tip: beat the market by tilting towards value and momentum assets, underweighting assets that are held for non-economic reasons, and taking advantage of having a long time horizon.

1. Expecting overly high returns from equities

Some expect to earn 7-10% per year from US equities (i.e. stocks and shares) over the next 10 years, whereas I expect returns to be more like 1-7%, with a best guess of about 4% (with lower expectations in the short-run, and higher expectations in longer time horizons).

High returns are usually justified on the basis of past performance. Over the long-run, US equities have returned about 8% per year. However, I don’t think extrapolation from past returns is particularly robust, because economic conditions can change, and realised returns – even over 10-20 year periods – have varied a huge amount in the past. Moreover my impression is that the significant majority of asset allocation experts expect returns to be lower going forward.

If you decompose the return of equities into their underlying factors, it’s hard to see how 8% returns will be possible. If you buy the S&P500 today and sell in ten years, your total return is captured by:

• The change in price between now and ten years time
• The cash you receive through dividends and share buybacks

You can split the first factor into:

• The growth in earnings
• The change in the earnings yield (P/E multiple)

Right now:

• The dividend yield is about 2%
• Buybacks get you a further 3%

So that gets you to a return of 5%.

Earnings growth historically has been 1.5% real, or about 3.5% nominal. However, it’s difficult to expect the same level of growth going forward. First, right now 100% of earnings are being paid out as dividends or buybacks, and nothing invested in growth, so it’s difficult to see where growth will come from (alternatively assume 1.5% growth in line with history, but reduce your expected long run rate of buybacks – overall you’d expect everything to add up roughly to the earning yield, which is about 5%). Second, profit margins are at record highs, and I think that means they’re more likely to go down rather than up, though the extent to which profit margins are mean reverting is debated.

If you assume 0% real growth, then you get 2% nominal growth, giving a total return of 7%. And I think that’s more or less the upper bound.

The final term, valuations, looks negative. Valuations are high whatever metric you use, and valuations also tend to mean revert over long time periods. Here’s a bunch of metrics that are correlated 0.5-0.8 with 10yr equity returns, all of which are unusually high.

So, it seems valuations are more likely to contract than increase, and this means returns are likely to be below 7%.

The exact reduction that can be expected is debated. The Shiller PE currently stands at about 26. If it reverts to its long-run average of 16 in 10 years, then that will reduce returns by about 5% per year, driving expected returns down to 2%, or 0% real. That could form the bottom of the range, giving an overall range of something like 2-7% nominal returns.

How come this is so much lower than history? Valuations were lower in the past, so returns were higher.

How come this is consistent with the efficient market hypothesis? The EMH doesn’t imply that the expected returns from the market portfolio are constant over time, just that these changes can’t be systematically exploited by traders to earn above market returns.

What does this mean?

• The lower expected equity returns are, the more likely it is that giving now wins over giving later.
• You may need to save more for retirement. If you want $1m at age 65 (inflation-adjusted), then you need to compound $175,000 of savings for 30 years at 6% (real). However, if you only earn returns of 3%, you’ll need to save $410,000, which is 2.4 times as much.

Further reading:

• A survey of valuation methods, finding a range of 0.5-3.5% real
• A survey of approaches by John Hussman, finding expected real returns of 1%, and a critique of these methods, and a partial response to that critique.
• A ‘bottom up’ estimate of equity returns finding 5% real
• Total equity holdings as a predictor, finding 5% real

Overall there’s substantial problems with all these methods, but I’ve found it hard to find anyone who expects returns over 7% in the next 10 years and has good reasons.

2. Not saving enough

Some in the community are lucky enough to have very high future earning potential, inherited wealth and a wealthy family who could support them if they hit hard times. In their case, it may be reasonable to donate almost all of your income, or live off very little and do direct work.

However, if you’re not in this situation, then you should save enough so that you have some personal runway to cover bad luck like getting sick for a long time and losing your job. You’re also going to need to save for retirement, and the earlier you save, the more time you have to compound, so the less you need to save. As a very rough rule of thumb, once you’re a couple of years into your career, you should consider saving something like 10-20% of your income until you’ve got enough for retirement. (More precisely, work out how much you’ll need, then play around with the numbers here to see how much you’ll need to save each year).

If you’re donating a lot, then you could save slightly less, because in an emergency, you could reduce your donations (unless the emergency involves losing your job).

3. Not being diversified enough

Many people I’ve spoken to are almost fully invested in US equities. I think the rationale for this is that equities have been the best returning asset historically, so there’s no reason to own anything else. Another rationale is that since you can’t beat the market, you should put everything into equities.

But US stocks do not equal “the market”. If you try to tally up all global financial assets, you get something like this:

• 18% US stocks
• 13% Foreign developed stocks
• 5% Foreign emerging stocks
• 20% Global corporate bonds
• 14% 30 year bonds
• 14% 10 year foreign bonds
• 2% TIPs
• 5% REITs
• 5% commodities
• 5% gold

This represents the truly agnostic portfolio. If you think you have no ability the beat the market, then this is the portfolio with the best risk-return. 100% US equities is a huge bet on just one asset.

From 1973 to 2013, a portfolio like this returned 9.9% per year. In comparison, stocks returned 10.2%. So you only gave up a tiny 0.3% to switch to this portfolio.

In return, you had far lower risk. The volatility of the 100% equity portfolio was 15.6%, whereas this diversified portfolio had a volatility of only 8%. The maximum drawdown was also only -27% compared to -51% with equities. The wide diversification also makes you less vulnerable to unforeseen tail risks.

The much lower volatility means you could have levered up 2x and ended up with the same amount of volatility and same drawdowns as equities, but returns that were twice as high, at 20% per year.

(It also had slightly higher returns than a 60/40 equity/bond portfolio (9.9% vs. 9.6%) but with volatility of 8% rather 10.5%. The returns and risk are also similar to the risk parity portfolio Ray Dalio recently recommended to Tony Robbins. And if you lever 1.4x, you get something that historically looks very similar to the true Bridgewater ‘All Weather’ portfolio).

(Source: Meb Faber’s book “Global Asset Allocation”)

Going forward can we expect a similar result? If US equities do unusually well compared to other assets, then the diversified portfolio is going to perform worse than 100% equities (as has happened in the last few years). But due to the far greater diversification, and so long as you have no strong reason to believe you can pick which assets will outperform, we should expect the global market portfolio to deliver the best ratio of risk to return. (If you believe you can predict that some assets have better risk-returns than others, then you should overweight those assets roughly in proportion to how much extra returns you expect, how uncorrelated the asset is with the rest of the portfolio, and how confident you are in your view).

If you can use leverage, then this means you can capture greater absolute returns too, because you can lever the global market portfolio until you have the same level as risk as whatever alternative you’re considering. (If you can’t use leverage, then it may be best to overweight whichever assets you believe have the highest expected returns, such as equities).

I also wouldn’t dial in, as per point 1, expected returns of around 10%. Bond and equity valuations are much higher today than 1973, so I’d expect something more like 4-5%.

You could make this portfolio yourself using ETFs (here’s a guide and here’s a simpler alternative), or this ETF does something similar though not exactly the same (this is not a recommendation to invest in this ETF and I have not investigated it thoroughly).

See some more discussion of the global market portfolio and another way to invest in it.

(Technically, the optimal portfolio would also be balanced with respect to your future income and other assets e.g. if you work in finance, then underexpose the rest of your portfolio to the finance sector (maybe even go short); if you’re going to inherit a house, then underexposure the rest of your portfolio to property. Otherwise you’ll be taking too much risk in these areas. However this is an advanced tip!).

4. Being undiversified and using leverage

There was a bunch of discussion recently about whether altruists should lever up their portfolio if they intend to donate the money later (leverage means taking on debt then investing that money). The conclusion was probably yes.

But the entire discussion was focused on whether to leverage US equities. Before you apply leverage, it makes a lot of sense to diversify your portfolio. Adding assets that are not perfectly correlated, especially during drawdowns, generally reduces risk much more than it reduces returns. This allows you to achieve a higher level of leverage, and overall higher returns.

As you can see from the earlier example, it makes much more sense to leverage the global portfolio than just US equities.

5. Taking too much risk

You should divide your portfolio into a personal component and an altruistic component. With the altruistic component, it probably makes sense to be pretty risk neutral, depending on which causes you support (and also see this). With high-risk money, if you don’t believe you can beat the market, then a leveraged investment in the global market portfolio is probably the best option. (Or if you can’t take on leverage, overweight towards whichever asset you expect to be highest return, probably equities).

However, with the personal component, you should definitely be risk-averse. It’s very important to save enough to be able to cover personal shocks such as illness, and have a modest income in retirement, but further wealth is much less important. However, I see many people with their portfolios almost 100% in equities, and I’m not sure they’ve properly set aside a risk-averse personal component.

A rough rule of thumb for buy and hold investors is to match the duration of your portfolio the time horizon of your investment. If you’re investing to retirement, then your time horizon is about 40 years. If you’re investing for emergency funds, then your time horizon is about zero, because you could need it at any moment. If you’re investing for a house, maybe your time horizon is about 10 years.

The duration of equities is about 20-50 years depending on their valuation. Right now, it’s towards the higher end in the US. So if you’re investing with a 10 year time horizon, and you only invest in US equities, the rule of thumb implies you should only have about 20% of your portfolio in equities and the rest in cash. Only if you’re investing with a 50+ year horizon should you be 100% in equities (and only if you truly have that time horizon, and won’t sell when the market falls!).

The global market portfolio above has a duration of about 25 years, and is overall much less risky, so you could have a substantially greater proportion invested (another way of seeing the returns to diversification). Matching duration would give you:

• 1 year time horizon: 4% in the global market portfolio and 96% in cash
• 10 year time horizon: 40% in the global market portfolio and 60% in cash
• 25 year time horizon: 100% in the global market portfolio
• 50 year time horizon: 200% in the global market portfolio (i.e. use 2x leverage)

Another way to come at this is volatility targeting. Take this risk aversion quiz. If it comes out high, target volatility of about 20%. If it comes out moderate, target 10%. If it comes out low, target 5%. If it comes out very low, target 1%.

• Very low risk: 12.5% in the global market portfolio and 82.5% in cash
• Low risk: 63% in the global market portfolio and 37% in cash
• Moderate risk: 120% in the global market portfolio
• High risk: 200-300% in the global market portfolio

To make your overall portfolio, add together the altruistic component to the personal component. e.g. if you want a low risk personal portfolio of $50,000 and an nearly risk neutral altruistic portfolio of $20,000 levered at 2x, then the overall exposure might be something like:

$50,000 x 0.63 + $20,000 x 2 = $71,500 in the global market portfolio and -$1,500 in cash

Update: added 30th Oct in response to discussion with Topher Hallquist. Note that I wouldn’t recommend having a 200% leveraged portfolio unless you have substantial experience with investing. You also need to be careful that you can access leverage cheaply enough that the expected returns are actually positive (e.g. buying a bond with a 2% yield to maturity on margin where the interest rate is 2% is pointless). Cheap leverage is often available through leveraged ETFs and semi-professional trading accounts like Interactive Brokers, but many retail accounts charge very high margin rates (over 5%) which would make it counterproductive to use leverage. If you can’t take on more than 100% leverage and have a high risk tolerance, then you have two options: (i) accept lower returns (ii) overweight towards the highest returning assets (likely equities) in exchange for much more risk.

6. (Advanced tip) Not beating the market

Most people I speak to are sold on the “expert common sense” view that amateur investors shouldn’t try to beat the market, and should instead invest in index funds. I basically agree with this view. However, if you’ve got a little more time to put into investing, I think it’s worth considering the idea of tilting your investments towards assets that have value (are cheap based on metrics like P/E and P/B), high momentum (have gone up in the last 12 months, are above their 200-day moving average) and low volatility. If you do this within equities, I think it’s possible to beat the market by a couple of percentage points.

These factors have been found to robustly outperform in equities and bonds across countries, variations in definition, over very long time periods, and when tested out of sample.

Value and momentum also seem to apply to asset class selection (as opposed to selections within asset classes)..

In part, the extra returns from these factors are earned by taking on extra systemic risk (e.g. value seems to underperform in deep bear markets), so are consistent with the efficient market hypothesis. However, I don’t think all the return comes from extra systemic risk, so there is a conflict with the EMH. I bite this bullet: I don’t think markets are completely efficient, and persistent behavioural biases create opportunities for systematic investors. This is of course not a widely held consensus.

Buying value stocks right now would involve loading up on Russian equities. If you feel averse to doing that, or averse to telling your partner you put the retirement money in the Russian stock market, then you can see why this bias persists.

How come the smart investors who know about these factors don’t ruthless exploit them and make loads of money until they’re all gone? There’s some evidence they’re going away in the most liquid, well researched parts of the market, but only slowly. The factors have been known about for decades, but there still just doesn’t seem to be enough capital systematically chasing these factors to fully remove them. I’d guess under 1% of the market is systematically tilting towards value, momentum and low volatility. Why? The outperformance of these factors is still not widely accepted among many investors, even professionals (and it wouldn’t be the first time researchers had discovered something important that isn’t widely implemented…); many investors aren’t interested in exploiting them, or can’t due to institutional constraints, and they’re not easy to exploit – you get a few extra percent of performance, but have to live with the risk of years of underperformance against the benchmark, which would cause most asset managers to lose their job. I expect the returns to these factors will go away over time, but I still expect excess returns over the next decade. See more discussion.

You can also beat the market for rational reasons if you have an unusually long time horizon or unusually high risk tolerance compared to the average investor.

1. You can buy illiquid assets, such as private equity and startups, which slightly outperform in compensation for the inconvenience of low liqudity.
2. You can do long-term contrarian market timing, in which you underinvest during good times, then invest heavily during crises. You earn excess returns by providing capital at the point when the rest of the market needs it most.

Finally, you can boost your returns by avoiding assets that other investors hold for non-economic reasons. For instance, UK pension funds are legally required to hold a large amount of long-term inflation-linked UK government bonds, so the yields on these bonds are unusually low. Avoiding these bonds can boost your returns. Some assets that might fall into this category include: government bonds in general (banks are required to hold them); corporate bonds, municipal bonds and real estate (often held due to favourable tax treatment, so less attractive if this tax relief doesn’t apply to you); gold (held by central banks without the aim of maximising returns).

If you want to be one step more sophisticated than the global market portfolio, then:

1. Tilt the portfolio towards asset classes that are especially cheap, low volatility and have high momentum (see here for estimates of asset valuations). One especially easy step would be to get equity exposure by buying a value equity ETF rather than a market cap weighted one. There’s now lots of these but unfortunately I haven’t investigated which are best. If you have more money and time, you could buy access to a screener and buy individual stocks yourself using a mechanical strategy. You could also hire a manager to do this for you, though this has a lot of challenges. AQR seems most looking into, but is only available to larger investors.
2. Also do this tilt across asset classes.
3. If you have a longer time horizon, then consider investing more in illiquid assets (e.g. startup equity, timberland, real estate) and not being 100% invested during good times so you can make extra investments during crises.
4. Underinvest in assets that others hold for non-economic reasons, such as government bonds, real estate and gold.

To be clear, all this section is an advanced tip – I don’t recommend doing it unless you’ve got substantial time to think about investing, enough money that the gains or worth the effort, and a good amount of discipline to stick to the strategy.

---

Update added 1st Nov 2015

Here’s a rough process for making your portfolio based on the above:

1. Decide what proportion of your savings are altruistic vs. personal.
2. For the personal component, work out your investment time horizon: when is the average time you expect to need the money? (You might find it helpful to divide it into several pots that all have their own time horizon).
3. Calculate the weighted average time horizon from both pots. You could use a 50 year time horizon for the altruistic component corresponding to high risk tolerance (or whatever level of risk you think is appropriate given your cause selection). So the weighted average is: 50x(altruistic proportion)+(personal time horizon)x(personal proportion).
4. Calculate your exposure. Roughly, take your time horizon and divide it by 25 (the duration of the global market portfolio). The percentage you get is your target exposure. If your target exposure is over 100%, we’ll come back to that later.
5. Invest your exposure target percentage into the global market portfolio. Put the remainder into cash. If you own a house, you effectively have a huge investment in property, which behaves a bit like an equity investment (with a bit more inflation protection). Don’t forget to include it in your portfolio. If most of your wealth is in a house, then you’ll want the rest of your portfolio to be conservative.
6. If your target is over 100%, you can use leverage cheaply and you’re an experienced investor, then consider leveraging the portfolio. If not, increase the weighting of equities.
7. Moderately advanced: Overweight towards your home country, since your future expenses will be in your home country currency. Add tilts to hedge to future income e.g. if you work in finance, underweight the finance sector.
8. Set it and leave it! The GMP doesn’t need to be rebalanced.
9. Advanced: add tilts to the portfolio for value, momentum and low volatility (either through security selection or asset selection or adding a long-short component) and away from assets owned for noneconomic reasons. In general, don’t do this unless you know what you’re doing! One very simple tilt you could consider, however, is replacing the equities with a value-weighted index e.g. replace VTI with VOE, though bear in mind this will slightly increase volatility and size of drawdowns.

Here’s a simplified version of the global market portfolio you can implement with ETFs (all $ denominated):

• 20% US stocks: VTI
• 21% International stocks: VXUS
• 20% US bonds: BND
• 27% International bonds: BNDX
• 2% Inflation-protected bonds: TIP
• 7% Commodities: DBC
• 3% Gold: GLD

Notes:

• I selected the ETFs from here and here. I believe their chosen on the basis of liquidity and low fees. The exception is the commodities ETF, where I chose one that aims to reduce losses from rolling contracts.
• Commodities are often not included in the GMP, but I included them because adding real assets should significantly reduce risk, because real assets hedge inflation shocks. If you think commodities have no risk premium, you could replace them with TIPs or remove them all together.
• I reduced gold from 5% to 3% because 5% is an aggressive estimate of gold’s proportion in the GMP, and it’s often held for noneconomic reasons.
• I removed the 5% in REITs, since these are already partially included in the equity indices, and it makes things simpler.