The Correct Cash Allocation For Your Retirement Portfolio Is Zero

Your cash bucket has a real return of minus 8.4%. It should be empty.

If you are in or nearing retirement, you have probably heard the lecture about a “cash bucket.”

Put one to two years of spending in a cash reserve, intone the experts. That way you aren’t forced to sell assets at an inopportune time.

“Fortify Your Retirement Against Market Whims,” recently in the New York Times, is typical of the genre. Sounds good. Why, with two years of cash in hand you can just shrug off market corrections.

I think this advice is bunk.

My view is that the correct amount of cash to have is the bare minimum—whatever you need for your checks not to bounce. As for the next two years of spending needs: Invest this money the same way you invest the rest of your portfolio.

To see what is wrong with the cash-as-buffer advice, try asking these experts a follow-up question: Just when am I supposed to replenish the bucket? Why should I expect that date to be any more opportune a time to sell?

Lying behind the specious appeal of a cash bucket is the trading fallacy. This fallacy is the notion that you can improve your lot in life by cleverly timing your buys and sells. The fallacy allows many a stockbroker to live well. It explains Robinhood. It explains most of the revenue at Coinbase Global, the crypto casino.

The trading fallacy has cash-bucket fans implicitly assuming that, since they “won’t be forced to sell at an inopportune time,” their sales will take place at opportune times.

Is there really any way to detect, in advance, whether today or a year from now is the more opportune time to cash out a stock position? If there were, nobody would need to fret about portfolio construction, or even to work for a living. Why, we could all buy at the lows and sell at the highs and live off the capital gains.

I think keeping cash at zero, or fairly close to zero, should be your goal. Don’t sell off portfolio assets now in anticipation of bills to be paid a year from now. Sell close to the time when you need the money.

What about coming up with money for an emergency? You need to replace a car, or your roof, when you weren’t expecting to.

Think this through. It’s not cash you need, it’s liquidity.

The usual settlement time for stock sales is two business days, for mutual funds and U.S. Treasury securities one day. There are times when you need money faster than that, such as if you are in a jail cell trying to make bail, but they are uncommon.

If you have a pile of financial assets, and presumably you do if you are debating how to structure a retirement portfolio, then you have plenty of liquidity. You do not need to have a percentage of your savings in a bank account or a brokerage sweep account earning 0.1%.

What about risk? Do you need a cash allocation in order to reduce the volatility of your portfolio? No, you don’t. There are better ways to reduce risk.

The problem with cash as a risk reducer is that it is a guaranteed loser. The nominal yield, which is what that bank or broker is quoting, is pretty close to zero. Subtract inflation, which came in at 8.5% recently and will probably average at least 3% over the next several years, and you have a real return deep into negative territory.

Here are four good ways to reduce the risk of a portfolio.

#1. Move from stocks to bonds.

A traditional mix is 60% stocks, 40% bonds. Maybe that’s too hazardous for you. Note that the Vanguard Target Retirement 2020 Fund, aimed at conservative investors just beginning retirement, is only 46% invested in stocks.

#2. Move from corporate bonds to Treasury bonds.

Corporate bonds pay better but they’re riskier. Some go bust.

#3. Move from long-term bonds to short-term bonds.

Long-term Treasury bonds have gotten hammered lately. The Vanguard Long-Term Treasury ETF has had a -20.5% cumulative return since the start of last year, per Morningstar data. If that’s too scary for you, shorten your bond maturities.

#4. Move from nominal bonds to real bonds.

Treasury Inflation Protected Securities due ten years from now are priced to deliver a real yield just a hair under 0%. Buy one of these things and you eliminate any uncertainty about the purchasing power of assets you’re setting aside now to spend in 2032. The price of the bond will fluctuate between now and then, but you know exactly where you will end up if you hang on until the bond is redeemed.

The problem with all of these risk reduction moves is that they reduce your expected return. I would assign to a 60/40 mix of stocks and corporate bonds an expected real return of 2% a year. Reducing risk means bringing that 2% number down.

In the extreme risk-reduction portfolio, a ladder of TIPS with maturities spread over your retirement, you push the real return down to 0% or thereabouts. For the very risk-averse, that’s not a bad idea. A zero real return is a lot better than the real return on cash, which is currently -8.4%.

Settle on a level of risk that you are comfortable with and stick to it. Rebalance occasionally, by selectively cashing out from the allocation that is overweight, but harbor no illusion that rebalancing will deliver timing gains or boost your expected return.