Using mutual funds and ETFs for short-term savings (1 year)

In investing, we make personal choices about how much risk we are willing to take if we think we can get higher return in exchange.  compares historical performance of similar funds that are appropriate for short-term savings.

Volatility--fluctuations in value--are an important aspect of risk. In financial economics, volatility is often measured by a statistical value called the standard deviation, represented by the Greek letter sigma (σ). The standard deviation of mutual funds can be found on sites like Morningstar or PortfolioVisualizer. However, it is hard to relate this number to a practical question, like "how much risk would I be taking by putting money I need a year from now into a short-term bond fund instead of a money market fund?" This article tries to help with this kind of question.

Overview
In this article, we calculate some other measures of risk by calculating what would have happened to an investment of $10,000, held for short periods of time, in a mutual fund or exchange-traded fund (ETF). You might do this in hope of making more than you would have made in a money market mutual fund, and accepting the possibility of sometimes making less. The possibility of making less is depicted in two ways:

1) What percentage of the time would you have made less than in a money market fund? We chose VMMXX, the Vanguard Prime Money Market Fund, because it is one of the oldest money market fund and gave us a long period of comparison. We also show what percentage of the time you would have actually lost money--ended up with less than $10,000 at the end of the time period--and what percentage of time you would have failed to keep up with inflation.

2) It's important to know not just when a fund or ETF underperformed a money market mutual fund, but by how much. In the case of short-term funds, the losses, when they did occur, were so small that some might call them negligible To quantify this, we present two numbers. One is the average loss that occurred in those periods in which losses did occur; and the other is the largest loss that ever occurred within the body of data that was used.

Notice that a high probability of underperforming may not be important at all. If two investments have the same return and volatility, but just fluctuate randomly with respect to each other, you would expect each of them to underperform the other about 50% of the time.

We have chosen to present data going back for as far as Morningstar has data. As a result, there are two caveats. First, the data range shown for each fund is different and results for different funds cannot be directly compared. For example, the Vanguard Short-Term Treasury Fund, VFISX, beat VMMXX by an average of $117.10 while the Vanguard Short-Term Treasury Index fund, VGSH, only beat it by $33.17. But this is almost entirely due to VGSH's inception in 2009 versus VFISX's inception in 1991. If we restrict our view of VFISX to the same years as VGSH, the benefit for VFISX was only $44.05. Second, this underlines the problem with all historical data, which is that the present time--and thus the short-term future--may be quite different from the historical averages.

Median and average (mean)
We show both the average and the median for some key values. A reader requested this. We have a scattered group of values: the return from an investment over many different periods of time. To represent the center of such a group with a single number, we often calculate and average or mean. In the case here, when we are looking, for example, at losses, and only seeing "the tail of the distribution," the average may give the wrong impression because it's influenced by extreme values. The median is another measure, and represents the value that is halfway based on counting. If the median loss of an investment is $600, it means you had a 50% chance of losing more over $600 and a 50% chance of losing less than $600.

Holding bond funds for the average "duration" of the fund
A "duration" is a calculation based on bond math that gives a number of years, related to the bond's term but shorter. One interpretation of the duration is that it is the "point of indifference." It is too well known that when interest rates rise, bond values fall. What is less well known is that if a bond is held for its duration, at the end of that period of time it will have the same value regardless of what interest rates do, provided the yield curve moves in parallel. This does not apply to a bond fund because the bonds in a fund are all different "ages" at any given time, and because the yield curve doesn't move in parallel. So there are no guarantees, but, nevertheless, the duration tells us a holding period over which the fund is unlikely to lose money.

This leads us to expect, for example, that "short-term" bond funds--like the Vanguard Short-Term Bond Index Fund, with a duration of 2.7 years--will sometimes lose money over periods of 6 or 12 months, but rarely lose money over periods of 5 years. This expectation is borne out by history.

A bond fund may not lose money, in the sense of dollars, yet lose value to inflation.