Before Alan Greenspan was a Fed chair, he was an academic economist, and some of his early work looked at the "wealth effect": when the stock market goes up, people are richer, and because they feel richer, they spend more. But the degree to which this effect is strictly the product of feeling richer is hard to measure because the market goes up for a reason: some combination of higher growth, lower rates, and perhaps just some invisible improvement in sentiment. So you'd expect there to be a positive correlation between consumer spending and the stock market even if the market had no causal impact at all.

Are we talking about the right wealth effect when we talk about stocks? Stock prices are easy to measure, and shares can be trivially converted into cash. But that's a lazy shortcut: equities are disproportionately owned by people with a low marginal propensity to spend. Of course, the very rich do spend plenty—the mansion/penthouse business seems to be lively in all sorts of macroeconomic conditions, yachts exist, etc. (Among the super-rich, the biggest-ticket items are extramarital affairs and visits to Epstein's island, but those aren’t fixed-price products, those prices are negotiated as a share of net worth.)

The biggest investment normal people make is in housing, and that does, indeed, drive a wealth effect. A billionaire who makes a few extra million when the market's up one day will not alter their consumption behavior much in response. Someone who bought a modest house in a nice area in the 90s and who discovers that, having paid off their mortgage, they own an asset that makes them a millionaire probably will change their self-conception a bit in response, and perhaps their spending, too.

But the real channel for the wealth effect is the more precarious category of people who own a home, have borrowed against it, and aren't averse to borrowing more. For a while in the mid-2000s, the total value of cash-out refinancings was running at ~3% of consumer spending, and this paper gets to a similar number for the spending impact, pegging it at 4.3% of GDP at the peak of the housing bull market.

For policymakers, this helps explain the mystery of why monetary policy works. A 25 basis point change in rates ripples through the economy, immediately affecting floating-rate obligations, gradually affecting the rates banks pay on deposits, and only hitting fixed-rate debt when people decide that the difference of 25 basis points is evidence that they really ought to borrow. The big exception is housing, where people can and do rapidly refinance when rates decline, and then promptly adjust their spending in response.

So some of the impact of rates is going to come through borrowers' disposable incomes, and the borrowers who see the relatively biggest impact will be the ones who are most financially stressed. But appreciating asset prices can be particularly pernicious for someone in that situation: for many homeowners in the 2000s, it was easy to get accustomed to having 5-10% more collateral each year, converting it into cash, and spending it. One way of looking at this is that it made the US's economic accounting less trustworthy: most economic activity was real, but the growth in that activity was increasingly based on people opting for a negative savings rate excluding asset appreciation.

That can't last forever, but it can go on for a long time: all that economic activity creates more jobs, and if the spending is directed into the right sectors, it, too, will lead to higher spending elsewhere.

All this ties back to an argument this newsletter has made in the past: the overwhelmingly most important impact most economic policies have is on the long-run equilibrium they establish for the creation and consumption of real resources. To be more specific: high and rising wealth inequality can be coupled with lower and shrinking consumption inequality if the richest increasingly don't spend their money. This was one reason for the slow economic recovery in the post-2000s period: there was actually more room for expansionary fiscal policy because so much upside accrued to people who were saving and compounding, rather than spending. That makes it the mirror image of the early and mid 2000s, where growth was slow, but would have been even slower if there hadn't been so much price appreciation in the typical middle-class portfolio, coupled with easy financial conditions for turning that into spending.

But even if the wealth effect is mostly a phenomenon that burns the middle class, it's at least a bit visible everywhere: ski resorts usually see better results when the market is up; even if their customers aren't literally selling a couple dozen shares of SPY to pay for the trip, they feel more like relaxing and less like skimping when the market's doing well. And there is one big, glaring exception to the general rule that rich people don't consume a large share of their income or their incremental wealth: back when Elon Musk was both the richest person alive and the best-paid CEO in history, he managed to blow 20%+ of his net worth on an addictive mobile app. The rich are not so different, after all.

Read More in The Diff

We’ve looked at feedback loops in markets, of which this is an example, many times. Some highlights:

• We’ve written about the mini wealth effect in tech, where rising stock prices make companies and employees feel more flush and behave more ambitiously.

• Central banks fundamentally transform land into liquidity ($).

• We’ve asked why there are so many equities fortunes compared to bonds and other assets.

• And we’ve covered some categories of extreme wealth that tend to be self-depleting ($).

Join the discussion!