Time to Rethink Excess Savings

Together the increased income and diminished spending resulted in quite a shift in household finances across all income groups. Whether you call it "excess" savings or pandemic-era savings, the windfall, along with frothy equity market performance in 2020 and 2021 culminated in a golden era of cash-rich consumers. The price-taking mindset of these flush consumers made a mockery of any finger wagging economist braying on about elasticity of demand, the idea that consumers will pull back spending as prices go up.

More recently as that excess cash has dried up, consumers have become a bit more choosy and price sensitive at least in some areas. So how much liquidity do consumers have? This has been a key question in the outlook for personal spending. Up until last week, we would have told you our homebrewed measure of excess savings suggested about five months of spending power left, getting us till the end of this year. But benchmark revisions to the BEA’s national accounts muffle these conclusions.

Early on, we were among the first to come up with a way of measuring that excess cash². We were encouraged that our straightforward approach to tallying excess savings found favor among other leading economists both in the banking sector and academia. We no longer feel that this approach is the best way to measure excess liquidity.

Our first clue should have been the annual revisions in September of last year. As we wrote at the time, the BEA’s annual revisions wiped roughly half a trillion dollars in savings off the previously published peak in excess savings³.

The benchmark revisions to the National Accounts last week caused some pretty big changes to our measure of excess savings again. Essentially, the data now suggest there is $1.1T remaining in excess savings through August, whereas the previous data suggested only about $339B remaining through July. This all comes down to our pre-pandemic baseline shifting from a 9.1% saving rate in January 2020 down to 7.2%. Since our baseline of comparison is now lower, the excess today looks higher because we are not as far from that base as we were. We have thus “spent it down” at a slower rate. The data now suggest there is more than three times the amount the pre-revised data suggested (~$340B) through July. The difference is clear in Figures 2 and 3.

Are consumers in better shape than pre-benchmark estimates? At the margin, perhaps yes. But the bigger point in our view is that it is not sensible to hang your hat on an estimate subject to such comically large revisions. Households still have excess savings upon which to fund spending and that pool of money is getting smaller. This insight need not rely upon data subject to such massive revisions. Instead, simply look at household checking & saving deposits relative to trend (Figure 4). This is a bit clearer and suggests households still have excess liquidity, though it is declining and excess cash does look to be concentrated in higher-income groups who have a lower propensity to spend it (Figure 5).

To be clear, we are not finding fault with the BEA’s data collection. Even when we implement others’ measures of excess savings, it shows a much longer runway than previously thought prior to the recent revisions. For example, the Federal Reserve Bank of San Francisco defines excess savings as the difference in actual saving and what is implied by the trend four-years leading into the pandemic⁴. When we replicate this calculation, excess saving goes from being run down in this September pre-revision, to suggesting households now have about ten months of spending power left before excess liquidity gets back to trend.

Ultimately, we will be moving away from these excess savings figures and will no longer make references to “how long the excess savings will last” as we think that question misses the mark. For many households, the excess savings is long gone. For others, especially wealthy ones, it can last much longer. If you don’t draw the money down, it can last indefinitely. Instead, we will consider the cash position of households by looking at savings and checking account balances relative to the linear trend prior to 2020.

Whatever the measure, excess liquidity will be less of a driver of spending going forward. Combine this with the fact that credit has become more expensive and harder to find and you are left with only real personal income as the sustaining driver of consumer spending. If the labor market retreats as we have in our forecast, that would clearly limit the capacity of consumer spending and would contribute to the mild contraction in Personal Consumption Expenditures that we have in our forecast.

Endnotes

1 - Aladangady, Aditya, David Cho, Laura Feiveson, and Eugenio Pinto (2022). "Excess Savings during the COVID-19 Pandemic," FEDS Notes. Washington: Board of Governors of the Federal Reserve System, October 21, 2022, https://doi.org/10.17016/2380-7172.3223. Also see "Consumer Staying Power by Income Group" November 2022. (Return)

2 - See "Flush: What Record Savings Mean for Spending" (June 2020) and "The Calm After the Storm: Sustainability Questions" (October 2020). (Return)

3 - See the U.S. Review section of "Weekly Economic & Financial Commentary" (September 30, 2022). (Return)

4 - "Excess No More: Dwindling Pandemic Saving." Federal Reserve Bank of San Francisco. August 2023. (Return)