Treating cash as a long-term investment has been historically a money-losing proposition.
This is fascinating because, while it strikes me as an obvious fallacy, it is presented as self evident, and he was sure enough to commit it to print (it’s quoted from his otherwise surely excellent book). It is common: I’ve more than once failed to convince people during a casual conversation that cash might be a fair long term investment.
But what happens if we look at the data? It’s not as if it’s hard to obtain, historical interest and inflation rates are some of most easily available economic data around, and many people have seen the times series many times and have a feel for where the nominal numbers have been in the past few decades.
So, here we have the real rate of return on cash from 1955 for the US dollar, sourced from Fred, as the Fed funds rates less CPI for inflation:
So we can see it’s been positive most of the time, it spikes down on occasion but usually hasn’t stayed negative very long. This can’t really produce negative long term returns very often, which we see below by showing the trailing 10 year returns, which is the return obtained from holding cash in an instrument that pays the Fed fund rates for 10 years and letting the interest compound. The numbers are shown annualised:
So we can see that apart from the immediate past, there’s only one instance in the past half century where holding cash for 10 years has had a — very slightly! — negative real return, circa 1980, at the worse time of the inflation period. For an instrument that is nominal-based, to work well even in a strongly inflationary period is quite an achievement. Not that it is particularly surprising either, interest rates do tend to be high, of course, during inflation periods, as economically it takes special circumstances to allow negative real returns to go far or last a long time.
The mean of all these periods is 1.8% which is not half bad and commensurate with typical historical returns on index linked bonds. This is consistent with long term efficient markets: if cash didn’t return something close to the long term risk free rate, few people would stay in cash for more than their petty cash and transaction requirements, which we know empirically is not the case.
I expect the result to hold for most other currencies, I just picked the dollar because it is familiar and the data was easy to find.
It may not remain the case in the future, but the claim was for history. If cash starts reliably losing money in the long (real) term from now on, which I guess is possible, it will be a fundamental regime change, a break with history.
It may be a “money losing proposition”, some of the time, relative to more racy investments, but rarely absolutely. In addition to the limited downside, in both real and nominal terms, you get no intermediary fees to pay, infinite liquidity, a somewhat pointless but still nice to have nominal capital guarantee, and historically small but steady real returns.
Is it a bad long term investment? It depends on circumstances of course, but it’s not easy to find an instrument that returns well above 2% real, and it’s even harder in risk adjusted terms.
For retail investors, the headwinds are even stronger, because there are so many pitfalls to non-cash investing (layers of fees, transaction costs, psychological biases, etc) and a retail saver can often beat the central bank rate while staying in cash: savings products are often available at loss leader rates from banks to entice new custom, and the simple strategy of moving your money every year or so to follow these offers is likely to achieve 1-2 percentage points above cash. When the consensus estimate for the equity premium is 3-4%, doing better than that is hard, and the sorry truth is that many individual savers would be better off if the only financial investment available to them was cash, even in the long term.
It is, however, extremely boring.
Disclosure: long cash.