The evolution of Target-Date funds

If you landed on this blog, you probably know what target-date funds are and why they are one of the preferred instruments to save for retirement. They are designed to be a simple, one-stop solution for savers. The concept is straightforward: pick the fund that matches your desired retirement date, put as much as you can into the fund and then forget about it until your freedom day comes.

In this post I would not delve into the “basics” of the instrument, it is not meant to be a beginners guide; what I want to discuss are aspects that I find interesting, how we got here, and maybe some pain points: the instrument is great but not perfect.

The risk profile of a Target-Date fund is not constant, its progression can be described in three phases: accumulation, glidepath and retirement.

Let’s look at each one of them.

The Accumulation Phase

This is the part of an investor’s lifecycle during which investors set aside money for their retirement. During this phase, cash flows are positive and contributions are going in. The fund is c100% invested in risky assets like stocks with the goal to maximize investment returns.

Recently, I read a paper that tries to prove that while high returns demand high-risk tolerance, when you have a long horizon high-risk tolerance is not required…at least if you live in Excel 😉 In the UK, but I guess in most other jurisdictions, you cannot touch your personal pension savings until a certain age, even if you default on your mortgage and are kicked out on the street. This means we should set-and-forget this investment; yet, we still freak out if the market crashes while we still have to wait 20 years before we can touch those funds. I say “we” because I do and most likely you do too, otherwise those principal-protected investments would not have the success they have.

After the doc.com bubble, a younger me was not convinced at all by the concept of “stocks for the long run”; the first FIRE generation, the one that saw stocks, and in particular tech stocks, go up in a straight line for like 11 years was probably pulled in the other direction and now is the day of reckoning. Doesn’t matter the origin, this fear of stocks never recovering is here to stay…because that’s where stock returns come from.

You probably have already seen these data somewhere, it takes a 5 second Google search:

There is no 20-year period in history where an investor lost money investing in the S&P, at least in nominal terms. If you know that you have to wait to get your savings, whatever volatility happens today should not be your concern. But:

• look at the debate around private equity and if investors should get an illiquidity premium or discount. Not being able to peek at the value of your holdings every 5 minutes (or being harassed by news channels) has value
• stocks will continue to grow in the future as they did in the past? This is a reasonable assumption…no, not that they will have the same returns as in the past but something closer, yes
• there are markets (Russia anyone?) where the above assumption has not been true. Investors need broad diversification, for example, because the best market of the past, the US, will not necessarily be the best in the future
• tangent to the previous point, there is a widespread propensity to offer a higher-than-benchmark home country index exposure to savers. That’s wrong for me, for one because we are way past the stage where, I do not know, the Italian stock index represents the Italian economy. That correlation the pension manager thinks they are offering to their clients is not there, and it’s even less where it matters the most, on the upside.

Portfolio risk is traditionally evaluated using a 1-year holding period, think about modern portfolio theory: optimal portfolios provide the greatest average expected utility from consumption after one year. If risk increases proportionally with the holding period, an investor’s optimal portfolio will not change. If risk does not increase proportionally, the welfare-maximizing portfolio is different. Indeed, it has been demonstrated that stock returns exhibit mean reversion and are not independent and identically distributed over time. This lowers the long-run holding period risk of equities and results in the Sharpe Ratio maximizing portfolios with higher equity shares for longer time horizons.

The research paper refers to this as “time diversification”. Time diversification assumes that the risk-adjusted performance of equities increases with longer time horizons. Assembling optimal portfolios involves consideration of not only investor risk tolerance, as suggested by modern portfolio theory, but also the distance between investment and consumption period.

The authors incorporate an autoregressive model in their Monte Carlo simulations, which leads to optimizations that better account for the superior performance of equities over long-run investment periods. The model shows what we see empirically: stocks do not only have a higher average return compared to other asset classes, those returns get more stable/less volatile the longer the investment horizon.

From the paper:

In Panel B, we see very little evidence of time diversification when returns are simulated without an autoregressive process – the optimal portfolio equity allocation does not change substantively across holding periods leading to the conclusion that holding equities over longer time periods does not reduce their relative risk. Panel A simulates optimal portfolios using the AR(5) model using coefficients drawn from historical U.S. data. The simulation does a much better job capturing the historical relationship exhibited by U.S., although optimal equity allocations are slightly less aggressive than historical estimates. An investor with an average level of risk aversion would allocate nearly double the allocation to equities for a long-run consumption goal versus a short-term investment. Differences between simulated and actual optimal portfolios may be attributed to the simplicity of the AR(5) model. For example, we assume that returns are normally distributed (since the error terms are normally distributed), while equity returns generally have non-normal characteristics.

This is the less controversial phase of Target-Date funds: behavioral aspects aside, pretty much everyone is on-board for a long time with the fact that if an investor wants to increase the probability of meeting their long-term goals, stocks are the place to go.

The Glidepath Phase

A glidepath is an investment roadmap that a Target-Date fund uses to take the investor from the accumulation phase all the way into retirement; along the way, the glidepath maps out a mix of stocks, bonds and other investments that is appropriate based on the target date. As time goes on, the glidepath automatically reduces the stock mix and adds in more conservative investments designed to preserve savings. As it gets closer to the fund’s target date, the glidepath arrives at an investment mix that’s designed to better support the last phase of the investor’s journey, retirement.

The glidepath typically starts 25 years before retirement. Why so long before? That’s the time that it takes to recoup a big market loss, according to (mainly) what happened in the past. The glidepath is basically a strategy designed to avoid sequencing risk. If an investor’s portfolio is 100% in stocks and the day before retirement the market drops 30%, the investor runs into the serious risk of not recouping ever those losses, even if the market bounces back, because they are ‘selling low’ to fund their retirement.

If I’m trying to develop the best strategy on WHEN you should de-risk your portfolio before retirement, the only data point of significance is WHEN the catastrophic market drop happens. The best time to de-risk your portfolio is right before that significant market correction, you are always better off taking lots of risk until a second before the correction happens. The glidepath is a great ‘general’ strategy for all the possible future paths investors will face. Is that strategy still optimal for the specific cohort of investors that is going to retire in 2035 and living in today’s market environment? In other words, should the 2035 Target-Date fund have a different glidepath, specific to what we know today, compared to the one of a fund that is still in the accumulation phase, like the 2065 TD fund?

This question is living rent-free in my mind for a while.

How often do stock markets correct? Michael Batnick wrote a post some years ago and I will steal some pics from him:

Again, consider that this data refers to the best stock market, not the average one, and remember that figures are not adjusted for dividends, so drawdowns reversed faster / the risk of selling while in a loss was lower.

If I am 10 years from retirement and today the market is down 30%, does the standard glidepath represent the best strategy for me, in terms of return maximization, or can I afford a higher allocation to stocks for the next decade? In this scenario, I am still contributing to the fund for a long time, would it make sense to buy more bonds when stocks are cheap? What about valuations? I know, trying to time the market based on those is a tricky proposition but 40% of stocks at P/E of 8 have a different risk than 40% of stocks at P/E of 25…no?

In reality, fund managers already have the ability to deviate, at least partially, from the standard path. This is what I found in the BlackRock LifePath Index Fund term sheet:

Although the asset allocation targets listed for the glide path are general, long-term targets, BFA may periodically adjust the proportion of equity index funds and fixed-income index funds in the Fund, based on an assessment of the current market conditions, the potential contribution of each asset class to the expected risk and return characteristics of the Fund, reallocations of Fund composition to reflect intra-year movement along the glide path and other factors. In general, such adjustments will be limited; however, BFA may determine that a greater degree of variation is warranted to protect the Fund or achieve its investment objective.

On a quarterly basis, the strategic allocation of each LifePath Index Fund is systematically adjusted to reflect the shareholders’ remaining investment time horizon and their updated strategic allocations. The Index is reviewed annually.

I looked for some real-life examples of the above, if BlackRock did any meaningful deviation in the recent past, but did not find anything specific. Then, it makes sense, given the run the stock market had since 2009. This is one area where I expect to see some innovation, maybe from small fund managers that want to try to beat the juggernauts of the sector like Vanguard.

The Retirement Phase

This is where most of the changes happened in the past and more are coming in the future.

If during the glidepath phase all savers targeting the same retirement date are affected by the same market conditions, the optimal retirement portfolio depends on each investor’s idiosyncratic situation. For one, different countries have different public pension systems: this annuity and how it is set to track inflation can leave retirees to be more or less aggressive with their savings.

The retiree has one principal goal: to not outlive their savings. There are several factors that contribute to the final outcome and only a few can be controlled by the pensioner. This creates a very complex problem to solve because there is no universal optimal strategy. There is also an inter-generational aspect to it: taking an adequate level of risk not only will protect the retiree from certain scenarios, like future high inflation, but will increase the probability of leaving part of the savings to children and grandchildren.

For these reasons, different money managers and index providers have designed different strategies (source):

As one of the proponents of a rising equity glidepath during retirement, the SMART Index is worth further analysis. Michael Kitces wrote a very interesting post on the topic: his research shows that despite the contrary nature of the strategy – allowing equity exposure to increase during retirement when conventional wisdom suggests it should decline as clients age – it turns out that a “rising equity glidepath” actually does improve retirement outcomes! If market returns are bad in the early years, a rising equity glidepath ensures that clients will dollar cost average into markets at cheaper and cheaper valuations; and if markets are good… well, clients won’t have a lot to worry about in retirement anyway (except perhaps how much excess money will be left over at the end of their life).

The standard period used in retirement portfolio research is 30 years. The more life expectancy grows or, who knows, the concept of FIRE (in some shape or form) becomes mainstream, the more the focus will shift towards ‘permanent portfolio’ solutions like the Cockroach I wrote about in the past. Going back to BlackRock as an example, their retirement asset allocation already includes 3.3% of real assets (Real Estate and commodities).

Conclusion

Ultimately, Target-Date funds try to solve in a single product what Meb Faber called the Get Rich and Stay Rich portfolios, from portfolios optimised by the distance between investment and consumption to portfolios that generates yearly income like endowments. It is funny to think they are perceived as very boring stuff while there is a lot of complexity, and interesting elements, if you just peek underneath.

Last thing. Looking at Target-Date fund mandates, it is not hard to realize that there is value in choosing a manager compared to the DIY version. Going back to the BlackRock example, the share class with the highest fee pays 44bps. So for 35bps, their fee minus the cost of the cheapest stock and bond trackers out there, you get below research embedded in your product:

Their higher fee is not 100% wasted money, especially if you do not have the time to stay updated with the latest advances in this space. (Despite it might appear so, I did not get paid by BlackRock to write this post. In fact, I only invest in a Vanguard TD fund).

What I am reading now:

Follow me on Twitter @nprotasoni