Good morning. In response to Friday’s letter on the persistent strength of the US economy — outside of the housing market — a reader made an obvious point, one we should have made ourselves. The economy’s resilience, though it sounds like good news, is going to make the Fed’s job a lot harder. It might have to tighten very aggressively to get demand below supply. But the Fed can hope for help from markets, as we discuss today. Email us: email@example.com and firstname.lastname@example.org.
The bear market in risk assets has been caused, in some part, by fear of a Fed-induced recession. But the causality can run the other way, too: if there is a recession, the bear market will probably be one of the causes.
Lower stock, bond and crypto prices make people feel poorer, so they spend less, making the economy smaller than it would otherwise be. For the Fed, at the moment, this is welcome news. They need growth to slow. But how much of a wealth effect can they expect? Might they get more help than they want?
Desmond Lachman, an economist and fellow at the American Enterprise Institute, pointed this out in response to last week’s letter about the economic slowdown:
The stock market rout of around 25 per cent has caused around $10tn in US household wealth to evaporate. In addition, at least $3tn in bond and $2tn in cryptocurrency wealth has been wiped out by the rout in those markets . . . On the assumption used by the Federal Reserve that a $1 sustained destruction in wealth leads to a 4-cent decline in consumption, if sustained, the recent loss in wealth could reduce consumption by almost 3 percentage points of GDP.
That last number struck me as pretty large, so I tried to reproduce it. Here is what I found:
According to the Federal Reserve’s distributional accounts, American households held $42.2tn in equities and mutual fund shares as of the end of 2021.
The Wilshire 5000, an index that captures virtually all publicly traded US shares, is down 25 per cent since year-end 2021 — a moment that happily coincides almost exactly with the top of the market. The Bloomberg long-term Treasury total return index is down by almost exactly the same amount. So making the simplifying assumptions that (a) Americans are exposed mainly to US stocks (b) Americans get their bond exposure through mutual funds, and (c) returns on that bond exposure has roughly tracked long-term Treasuries, we can assume that household portfolios have lost $10.6tn in value this year.
A paper from the Federal Reserve looking at fluctuations in household wealth and spending during the 1990s market boom finds that “groups of families whose portfolios were boosted the most by the exceptional stock market performance over the latter half of the 1990s are the same groups whose net saving flows fell the sharpest from 1995 through 2000,” and that “The [resulting] movements in net worth and saving are consistent with a wealth effect in the range of 3-1/2 to 5 cents on the dollar that applies to all families in the economy.”
So assume, conservatively, that every dollar lost in markets translates to 3.5 cents of lost spending. That comes to $370bn in lost spending, or about 2.6 per cent of consumer expenditures or 1.8 per cent of GDP.
The total market capitalisation of crypto assets has fallen from $2.9tn to $835bn, according to CoinMarketCap. This translates to another $73bn in lost spending, and another third of a per cent or so of GDP.
So for a first, conservative estimate, the bear market’s impact on spending could amount to a 2 per cent drag on GDP.
That’s a lot! Recall that in the first quarter GDP grew 2.6 per cent, once some odd fluctuations in inventories and net imports are stripped out. Forecaster consensus calls for GDP to grow 1.8 per cent for the rest of the year, with consumer spending growing a bit faster than that, with the basic pattern to persist for 2023. How much of a negative wealth effect is built into those forecasts I have no idea.
It’s a very rough and ready estimate, though. Sensitivity of spending to loss of wealth likely depends on hard-to-measure psychological factors. For example, it must matter how much people are counting on the wealth they have gained in markets, as opposed to thinking of it as a windfall. Did households treat their sudden crypto gains as “found money” rather than hard-earned savings, and therefore change their spending patterns less in response to their appearance and evaporation?
About 20 years ago, a Fed board member, Edward Gramlich, made the point that in theory, the impact of stock market wealth on consumption should depend on whether stock prices rose because profits rose, or because discount rates fell (or what amounts to the same thing, price/earnings multiples rose):
Suppose, for example, that stock prices increase because of a rise in expected profits, say from a spurt in productivity. An individual household that owns stocks will have higher wealth and will want to consume more, just as predicted by the wealth effect . . .
[if instead] stock prices increase because households are applying a lower discount rate to future profits [then whether] an individual household will want to consume more is unclear in this case. Intuitively, households are simply discounting the same stream of profits at a different rate; so it is not obvious that they are truly better off and should increase their consumption.
I would make sense of this point intuitively as follows. If my stocks have gone up because of strong economic growth, the gains make sense to me. I’m an owner of growing businesses in a growing economy. If they go up because of multiple expansion, that feels chancy. I trust the new wealth less, and might be less inclined to let it change my spending patterns, as a result.
I really am not sure if the source of market returns matters to spending or not. But it is a particularly important question right now.
During the 2009-2021 bull market, the S&P 500 rose 325 per cent (if you use year-end 2009 as the starting point). Earnings over the period grew 192 per cent. The rest of the gain was down to price/earnings ratios rising from 16 to 24. Much of the latter increase has been given back (the index is now on 18 times trailing earnings), even as earnings have hung in there, for now. How will household spending patterns respond to a shift in valuation without much shift in profits?
One good read
If you’re reading this newsletter, decent chance you care about the Fed. Most people don’t, though. Could that matter to inflation expectations?