The False Promise Of Target-Date Funds

February 14, 2014

In spite of the de-risking claim, risk from all investments, including those in earlier years, propagates forward to final wealth. It is doubtful that a predetermined schedule of risk targeting, ignoring information updates, is a good or even acceptable way to optimize retirement wealth through uncertain markets. To achieve its intended risk characteristics at retirement, a glide path must be followed to completion and cannot be adjusted midpath. Good investments depend on more than the number of years until retirement. New information must be recognized and acted upon to provide proper fiduciary service to investors. Through the savings period of a retirement plan, personal circumstances change, and it seems unwise to leave retirement assets unattended in a locked one-size-fits-all plan that completely ignores both market and investor.

A better default investment for long-term investors without opinions or views about markets may be a strategic diversified portfolio close to the composition of the market, such as a 60 percent equity and 40 percent fixed-income portfolio. Target-risk portfolios are more transparent about their risk exposures and offer expended possibilities of adding value by seeking improved diversification across risk factors and asset classes that may be unattainable in a glide path. The costs and implementation of funds for many target dates may put constraints on the ability to cross-sectionally diversify a portfolio.2

This article sets out to reveal the relationship between glide paths and the return distribution at the target date through a simulation experiment. Our hypothesis is that predetermined portfolios up to the retirement date provide scant advantages over other possible plans, given market uncertainty. Rigorous glide paths amount to bets on particular market outcomes. We hypothesize that the “best” glide path will depend strongly on any observed patterns in market returns, and therefore that a specific choice of glide path is a bet that market conditions will change accordingly.

To provide evidence that our simulations are consistent with market realities, we also analyze glide paths on historical return data; show that outcomes are time-dependent across glide paths; and that the choice of glide paths has far less impact on retirement wealth than the overall risk tolerance through the entire investment history. Indeed, the historical example suggests an investor in a typical glide path has both a higher standard deviation and a lower median wealth than a comparable fixed-risk investor.

The article is organized as follows: In the next section, we describe the main simulation experiment. Then we present some details about the methodology in computing the simulations and statistics. In the section after that, we present the simulation results and the results of our historical analysis. Next, we test the robustness of our conclusions to changes in some of the assumptions of the experiment, and address some other concerns about the experiment before discussing our conclusions. Additional technical analysis of the matching on standard deviation is included in the appendix.

The Simulation Experiment
Our simulation experiment illustrates how glide paths affect retirement wealth and how the selection of the best glide path relies on results that are unknowable a priori, offering a false sense of risk control to the investor.3 Of course, TDF products on the market are far more intricate and contain detailed glide paths for other features than just stock/bond ratios; nevertheless, we find the simple example presented here compelling. Adding complexity to the glide path only adds to the specificity of the assumed future performance of the markets.

Investor contributions are difficult to generalize. A typical investor may increase saving through the years before retirement as income allows. Since erratic contributions further hinder the value of a glide path, we assume a steady contribution, growing by a constant inflation rate, over a 40-year period. This would match a retirement investor saving, for example, from age 25 to 65, making the maximum matched contribution to a 401(k) plan over the course of a typical lifetime career.4



Find your next ETF

Reset All