Skip to main content


Saving and COVID-19

Saving and COVID-19

Record Saving is Protecting the Economy and Will Help Return Americans to Work

One unusual feature of the U.S. economy during the COVID-19 pandemic is a massive increase in saving. Saving nearly tripled over the first two quarters of 2020, from $1.59 trillion annualized in the first quarter to $4.69 trillion in the second.1 This was by far the biggest increase in modern history. While the magnitude of this change in the economy is striking, it should be lauded: the saved money is protecting the economy from painful disruptions, and furthermore, it will be instrumental in reconnecting millions of Americans to work after the pandemic subsides.

2020's Historic Rise in Saving

Source: Bureau of Economic Analysis, Gross Domestic Product, Table 8,

Saving often gets a bad rap, especially during recessions. Rather than save, the theory goes, people should spend their money immediately on goods and services in order to raise the incomes of their neighbors. During periods of higher unemployment, that spending is especially likely to create jobs, employing new workers in the production of the desired goods. This thesis is most widely attributed to John Maynard Keynes.2 However, as Keynes notes, elements of these ideas predate his lifetime.3  With an unemployment rate of 10.2 percent for July, the United States would seem to be in the situation where this theory holds.

However, as a means of analyzing the U.S. economy in 2020, this story is missing some relevant details: first, it has not been adapted for the unusual circumstances of the COVID-19 pandemic. Second, it elides the connection between saving and investment. Finally, one of the major concerns Keynes had about saving—that saved money would not raise the income of ones’ neighbors—has been addressed through other means.

After including these details and examining the data of 2020, the historic personal saving of Q2 2020 looks like a boon. It has been protecting the economy, lowering the cost of capital and providing financing to homeowners and firms in tough times. Furthermore, it will eventually reconnect millions of Americans to work; the savings will allow households to spend more on goods and services once it becomes safe to do so, and this will in turn draw unemployed workers back into the economy.

Optimal Market Spending is Lower Under COVID-19

Gross Domestic Product (GDP) fell from an annualized $21.6 trillion in the first quarter of 2020 to $19.4 trillion in the second quarter. This change is directly connected with the change in personal saving; people are forgoing the sort of spending counted in GDP, and are saving their personal income instead.4

Declines in GDP are generally considered undesirable, and with good reason: it is a measure of voluntary exchanges that are beneficial to both parties. If the voluntary exchanges are gone, so too are the benefits. Under normal circumstances, it is worthwhile to reverse a falling GDP.

However, GDP has some shortcomings as an overall measure of well-being, and they are particularly acute under the COVID-19 pandemic. The first is that GDP does not directly measure public health. The second is that GDP does not account for the value of work done at home.5 After accounting for these shortcomings, the drop in GDP is less undesirable than it would be under normal circumstances. Therefore, savers may be justified in withholding their spending despite the decline in GDP.

For example, consider how the economics of food have changed in 2020:

Food Spending: Less Dining Out, More Groceries, Lower Spending Overall

Source: U.S. Bureau of Economic Analysis, Gross Domestic Product, Table 3, 

Spending on grocery items surged to new all-time highs, while spending on food services, like restaurants, has plummeted. The latter effect is larger, so spending on food has decreased, and our GDP in the food sector has decreased. While this would be undesirable under normal circumstances, there are reasons to think it less so at the moment.

The first is that in limiting close-quarters contact with others, people are engaging in a kind of public health “production”6 and rationally responding to the threat of COVID-19. The second is that reduced market activity has corresponded with an increase in at-home activity.

Some tasks, like cooking food, plating it, cleaning dishware, and creating a dining ambience—have left the market economy temporarily. GDP has fallen by the value of those tasks. However, home production in those tasks, which is unmeasured by GDP, has likely risen, as people turn groceries into meals. Similar trends are happening throughout the economy: from cooking to cleaning to childcare, market activity has receded but the tasks are still getting done.

By all means, this is a downgrade from the status quo prior to the pandemic. It is better, for example, to have our cooking output concentrated among our better culinary talents, for the quality and the economies of scale. However, home production has some value, too.

But an immediate and complete return to the pre-virus status quo does not look likely . Given the menu of options available to Americans right now, the choice to save more, spend less, limit the spread of COVID-19, and engage in more at-home production is a reasonable one, even if it comes at some cost to market production.

Personal Income Rose in the Second Quarter

Another important facet of the COVID-19 economy is the role of public policy actions taken so far. They have increased disposable personal income, largely through transfer payments. Major components of personal income have changed as follows from the first quarter to the second, on an annualized basis:7

  • Employee compensation fell by $795 billion, reducing disposable personal income.
  • Proprietors’ income fell by $224 billion, reducing disposable personal income.
  • Tax collections fell by $148 billion, increasing disposable personal income.
  • Transfer receipts rose by $2.419 trillion, increasing disposable personal income.
  • Overall, disposable personal income rose by $1.535 trillion.

As mentioned above, one of the hypothetical drawbacks of saving during a recession is that forgoing spending may harm the incomes of others; one person’s spending is another person’s revenue. Should enough people forego spending, income as a whole will fall. One can see this effect in the drop in employee compensation and in proprietors’ income.

However, overall income continued on an upward trajectory. The effect of lower taxes and increased transfer receipts was greater than the effect of reduced consumer spending. The largest changes came from the Coronavirus Aid, Relief, and Economic Security (CARES) Act, though other laws also played a role. There is plenty of room for debate about the specifics and overall merits of these laws, which will not be addressed here. These laws are mentioned only because they raised personal income overall, a fact which is relevant to the Keynesian critique of saving.

There is another factor that mitigates the Keynesian demand-side critique of saving: many kinds of saving help finance spending elsewhere in the economy: for example, mortgage lending can help finance the housing market and spending by households. Investing in corporate bonds or stocks can make it easier for firms to raise money and incentivize them to spend. To the extent that participation in capital markets enables spending by firms or other individuals, it still provides income for others just as consumption does.

Below will discuss how these capital markets channels are unusually important during the COVID-19 pandemic, and why it is beneficial for savers to put so much money into them.

How Lower Mortgage Rates Have Helped the Economy

One area where savers contribute to the economy is the mortgage market. They accept some default risk (on loans that are not federally backed) and earn a risk premium—or higher rate of interest than one would find on risk-free debt—in exchange.

In recent months, mortgage rates have reached an all-time low.

Mortgage Rates Reached Historic Lows in 2020

Source: Freddie Mac, 30-Year Fixed Rate Mortgage Average in the United States [MORTGAGE30US], retrieved from FRED, Federal Reserve Bank of St. Louis;, September 28, 2020. 

This may be surprising, given that mortgage delinquencies have risen substantially, from 4.36 percent in the first quarter of 2020 to 8.22 percent in the second.8 Mortgage products have gotten worse, not better, from the lender’s perspective, but nonetheless they are being transacted at lower yields (or higher prices). While savers would certainly prefer a higher rate of interest on their lending, the extreme quantities of savings in the COVID-19 economy are forcing them to compete with each other on risk tolerance and offer more attractive rates to borrowers.

More attractive rates for borrowers has several helpful qualities. First, it may allow current borrowers to refinance at lower rates, reducing their monthly payments. Second, it keeps home prices stable; the lower the mortgage rate, the more one can pay for a house on a given income. Stability in home prices in turn keeps financial markets and the cyclically-sensitive construction industry on track.

Home Prices Remained Steady Through 2020

Source: S&P Dow Jones Indices LLC, S&P/Case-Shiller U.S. National Home Price Index [CSUSHPINSA], retrieved from FRED, Federal Reserve Bank of St. Louis;, September 29, 2020. 
Note: The Case-Shiller Index is a measure of house prices that uses a "repeat-sales" method, meaning it attempts to track the overall price level of houses by observing the changes in prices of homes that are sold multiple times.

The COVID-19 recession is entirely invisible in terms of home prices. Lower mortgage rates have allowed buyers to continue paying the pre-pandemic market prices for homes, and to do so with lower monthly payments than before.

Stable home prices are desirable from the perspective of financial stability. Houses are a common and valuable kind of collateral, and loans are secured against them. When collateralized loans reach a high loan-to-value ratio, defaults and other disruptions become more likely, as experience from the 2007-2009 housing crisis shows.

Perhaps even more importantly, stable home prices also help keep the construction industry on track. Given that their final products are still selling normally, construction firms can generally afford to spend as usual, and pay workers their usual wages.

After an Initial Pause, Housing Construction Bounced Back in Summer of 2020

Source: U.S. Census Bureau and U.S. Department of Housing and Urban Development, Housing Starts: Total: New Privately Owned Housing Units Started [HOUST], retrieved from FRED, Federal Reserve Bank of St. Louis;, September 29, 2020

Housing starts struggled in the early, unstable spring months of COVID-19, but as uncertainty receded, construction returned to pre-crisis levels, powered by the low mortgage rates.

High Valuations, Low Yields

Some features of the mortgage market also apply to corporate stocks and bonds. Just as in the mortgage market, savers are accepting lower long-run rates of return than before, even as the economy becomes more uncertain.

For example, as of August 21st, 2019, the S&P 500 index was trading at 22 times its trailing-12-month earnings. On the same day a year later, the S&P 500 was trading at 35 times its trailing-12 month-earnings, and 26 times its expected earnings for the coming year.9

Similarly, in bonds, yields have fallen. (Or alternatively, bond prices have risen.) For example, the Moody’s Seasoned Baa Corporate Bond Yield is a measure of the interest rate for moderately-risky bonds: neither the safest of investment-grade bonds, nor risky junk bonds. In other words, it is a fairly-representative example of the sort of rate at which many actual firms can borrow. The Seasoned Baa Corporate Bond Yield has fallen from 3.6 percent in February to 3.3 percent in August.

Corporations Have Historically-Low Cost of Capital in 2020

Source: Moody’s, Moody's Seasoned Baa Corporate Bond Yield [BAA], retrieved from FRED, Federal Reserve Bank of St. Louis;, September 28, 2020

Both debt and equity stakes in firms are transacted at higher prices, even as the underlying assets have worse future prospects due to COVID-19. Despite its counterintuitive nature, this market response is appropriate to the conditions at hand.

Asset prices are an exchange rate of sorts between present and future. One gives up present cash for claims on the future. Right now, present cash is plentiful, because saving has risen. Current purchasing power is less useful than usual; many normal economic pursuits are currently risky or even prohibited. Naturally, many people want to exchange present purchasing power for claims on the future. An asset rally does not necessarily signal optimism about the future; it can just signal an increased desire to defer consumption.

While this result is partly an accident of the peculiar circumstances of COVID-19, it happens to have desirable properties—in fact, very much the same desirable properties that the low mortgage rates have had for housing construction.

Suppose a company has a 5 percent cost of capital, and it is interested in a project that would cost $10 million in one-time expenses, paid to workers. The project would then return $500,000 a year in annual cash flows from then on. Under the company’s 5 percent cost of capital, the project can finance itself successfully. Now consider what happens if the economy takes a turn for the worse, and the same project is expected to return just $400,000 a year in cash flows. The project returns only 4 percent, and is no longer worthwhile if the 5 percent cost of capital remains the same.

There are two realistic options. The first is that the project could be canceled. The people it would have employed will have to search for other jobs, and the savers who would have funded it will have to search for other investments. The second option is that investors could accept a lower rate of return.

If this is an idiosyncratic problem with just the one project—and not a broader problem with the economy—the best solution for society at large is the first one. Workers and savers should coordinate their talents and resources on projects with greater overall returns.

However, imagine this is a systemic problem—a recession—with many projects suffering the same problem simultaneously. Then the first option isn’t so rosy; the workers will likely not find a different job, and the investors are unlikely to find equivalent or higher returns elsewhere. Better to just acknowledge the reality that the economy has taken a turn for the worse and have investors accept lower-yielding assets. Low interest rates are a signal to firms to go ahead and carry on through tough times, even if things are not as good as one had hoped for.

The investment process described above is abstract, simple, and mechanical; it may not describe the diversity and idiosyncrasies of actual corporate managers’ thought processes. However, most plausible behavioral responses run in the same direction: lower cost of capital encourages greater spending. For example, suppose management of a mature company buys back shares when they feel those shares are undervalued, but otherwise spends the cash on hand. Or suppose management of a startup decides to dilute their ownership stake by a particular percentage in their next funding round, and then spend that cash on expanding the company. In each case, the firm would end up spending more if valuations were higher and interest rates lower.

Empirically, the summer asset rally has been associated with a bounce back in orders for capital goods. Manufacturers’ new orders for non-aircraft capital goods have returned to $66 billion, just where they were in February.

After an Initial Pause, Capital Expenditures Bounced Back in Summer of 2020

Source: U.S. Census Bureau, Manufacturers' New Orders: Nondefense Capital Goods Excluding Aircraft [NEWORDER], retrieved from FRED, Federal Reserve Bank of St. Louis;, September 29, 2020. 

This chart is similar in its V-shaped recovery to the housing starts chart above, and is in fact driven by the same mechanism: heavy savings with realistic expectations about yield have enabled life to return to normal in construction and manufacturing.

Cash Flow Management and Organizational Capital

Firms use capital markets to raise cash now in exchange for claims on the future cash flows of the business. This practice is naturally most useful when a firm is expected to generate positive cash flows in the long term, but its near-term cash balances are declining or expected to decline.

Under normal circumstances, only some firms need access to capital markets. For example, new businesses that spend substantially on customer acquisition, research and development, or investing activities may generate negative cash flows for several years as they get off the ground. Some mature businesses may need outside cash to fund larger investments or acquisitions.

During the COVID-19 pandemic, though, there are likely to be many more firms than usual that need access to capital markets. There are many firms that were successful prior to the pandemic, and are expected to be successful afterwards, but temporarily have lost some of their revenue sources for public health reasons. These firms have the precise qualities where borrowing is most important: they have negative cash flow in the short term but positive cash flow in the long term. They are worth saving in the short run because they are good organizational arrangements in the long run. If there are deep capital markets—markets awash in saving, with a diversity of lenders and investment managers supplying different kinds of loans or equity injections—these firms can be “saved” entirely by private markets, even without any controversial public-sector “bailouts.”

Going to capital markets for additional funding will keep these firms organizationally intact, but dilute the stakes of current shareholders. This is a superior solution to dramatic, costly liquidations. Keeping firms relatively organizationally intact is likely to be important in terms of returning Americans to work. Work is a series of relationships; it is not just a frictionless spot market where people bloodlessly trade an undifferentiated commodity. People who work together develop an understanding of each other’s working styles and strengths, and even become friends. They become accustomed to a particular organizational system, and get good at working within it.10

Above, the examples of investment have been about homes and capital goods; however, corporations can also “invest” in worker retention during recessions, paying them more than the revenue their efforts bring in. This process, sometimes known as “labor hoarding,” may be rational and profit-maximizing in that it preserves organizational know-how, or it may be management electing to sacrifice some profitability for the sake of comity, loyalty to workers, and overall welfare. In either case, there is value in the approach, and it would be good for capital markets to consider helping firms take it. Then, when industries resume full operations, firms can simply “get the band back together” rather than starting from scratch.

Most negative-cash-flow firms in 2020 do not need dramatic shakeups, they just need time until the pandemic subsides. Deep capital markets can grant them that time.

Spending Again When Consumption is Safe

The most important feature of money saved in 2020 is the simplest: it prepares Americans to spend amply as the pandemic subsides. Annualized personal consumption expenditures will likely need to rise about $2 trillion from the second quarter trough, with a particularly large rebound in food services and accommodations, and recreational services.

Returning these industries to work is not only good for its own sake, but also because the workers in those industries are customers in other fields. For example, software developers may not be directly affected by COVID-19, but they are indirectly affected if their customers’ income falls. The engine of the economy is calibrated with the expectation that flight attendants, line chefs and box office managers have money to spend. In the absence of that money, it will suffer.

Once these individuals can return safely, it is best for them to return to work rapidly so they can earn and spend money. A speedy return to work, in turn, would best be accomplished with a surge of new orders: flight bookings, dining reservations and concert ticket sales. If firms are faced with high demand, they will quickly staff up to fill orders.

The super-charged $4.7 trillion in annualized saving from the second quarter—much higher than the $1.2 trillion of 2019—is critical to making this scenario a reality. The engine of the consumption economy is powerful, but it has been partially shut off. Savings are the stored energy in a battery that can be used to start the economic engine again.

One might note that the saving is more than sufficient to bring personal consumption expenditures back to trend. It may generate too-high inflation if it were spent all at once. However, there are good reasons to believe it will not be. Windfalls are often spent down over many periods, not all at once. Furthermore, a rise in interest rates could incentivize more consumption smoothing should inflation arise.


The historic and anomalous rise in saving in 2020 was a direct product of the COVID-19 pandemic and the associated policy responses. While saving has some drawbacks in normal recessions, those drawbacks did not apply in the second quarter of 2020; policy increased household income and more than made up for the loss of income from consumer spending. Furthermore, forgoing some consumption is a rational choice with public health benefits in the short run.

The strong supply of saving has led to high valuations but low expected rates of return. This financial environment has helped the housing market to continue functioning without disruption, and has allowed firms to return to ordering capital goods. This form of sacrifice—in which savers accept lower future returns—is one of the smoothest ways that markets handle recessions, and better than most alternatives.

Saving also has some benefits specific to this recession, as more firms than are typical will need to draw on capital markets. Many firms have negative cash flows in the present, but hold a wealth of valuable employee and coworker relationships and are profitable in the long run, suggesting they should be preserved. A strong supply of saving and robust capital markets help in that preservation. One recession-fighting measure sometimes used by governments is to extend emergency loans to business, either through legislation or “extraordinary measures” by central banks. However, these measures often result in concerns about corruption, political economy, or constitutionality. It is therefore much better if the private sector has the money and power to accomplish some of these goals on its own.

Finally, a strong supply of saving will help restart affected sectors of the economy when it is safe to do so. Ideally, for example, a family might have some money set aside now to take a vacation or go to a popular concert once it can be done without excessive public health risks. A flurry of new orders in such industries will reconnect millions to work.

The historic rise in saving in the second quarter of 2020 came partly by accident; it came from people deferring consumption for health reasons, not necessarily a coordinated strategy to increase the supply of saving. However, it has nonetheless had strong benefits and helped housing construction and manufacturing bounce back quickly. This suggests that robust saving helps solve some problems. Money in the hands of the American people—even if that money is saved—is a strong component of a recession-fighting strategy.

Alan Cole
Senior Economist

1 Bureau of Economic Analysis, Gross Domestic Product, Table 8,

2 Keynes, John Maynard, The General Theory of Employment, Interest, and Money (1936), Chapter 23

3 See Bernard Mandeville, The Fable of the Bees: or, Private Vices, Publick Benefits (1714), or, for a more modern telling, see Paul Krugman, Babysitting the Economy, (1998),

4 This description is simplified and partial-equilibrium. At the macroeconomic level, one individual’s saving may reduce income for others. However, personal income did not decline in 2020, which will be addressed further on.

5 There are good reasons not to include household production in GDP, since market production and household production have some important differences, and it is useful to have a measure of market production alone. For the Bureau of Economic Analysis’s explanation, see Bureau of Economic Analysis, “Why isn’t household production included in GDP?”, April 16, 2018,

6 The work of Elinor Ostrom covers extensively the idea that much output is produced by a combination of market and non-market work, coining the term “co-production.” In 2020, many individuals with jobs outside the healthcare sector are engaging in behaviors designed to help that sector. While the value of co-production is often unmeasured, common sense suggests it has value nonetheless.

7 Bureau of Economic Analysis, Gross Domestic Product, Table 8,

8 Mortgage Bankers Association, “Mortgage Delinquencies Spike in the Second Quarter of 2020”, August 17, 2020,

9 Birinyi Associates data retrieved from The Wall Street Journal, August 25, 2020

10 Becker, Gary, Investment in Human Capital: A Theoretical Analysis, The Journal of Political Economy Vol. LXX No. 5 Part 2, October 1962,

Latest News