According to a May 2020 CNN article, one consequence of the coronavirus pandemic is that “Americans are slashing their spending, hoarding cash and shrinking their credit card debt as they fear their jobs could disappear.” The article goes on to say that U.S. credit card debt has fallen by the largest percentage in thirty years while household savings rates have climbed to levels unseen since the 1980s.
While part of this increase in savings may be a temporary reaction to the lockdown, and consumer spending could very well pick up again quickly once conditions return to normal, there are many reasons to believe and some evidence to suggest that American households will cut back permanently on at least some future discretionary spending to have a higher savings buffer in case of future crises.
The idea that Americans want to consume less and save more might seem at first like a good thing, but it could put a severe damper on the consumer-driven U.S. economy as a whole: after all, one person’s spending is another person’s income. So even if individual Americans strive to save more money, that may not be enough to drive up the collective American savings rate, which can only rise under very specific—and improbable—conditions. Put differently, if some American households decide to save more, the country’s overall savings rate will not rise unless some other (unlikely) adjustments or tradeoffs elsewhere in the economy allow that to happen.
What’s more, higher household savings isn’t always a good thing for the broader economy. It could be positive or negative for the United States depending on what the country’s pandemic-era economic adjustment looks like. Because total U.S. investment is by definition equal to total American savings (that is, the savings of households, businesses and the government) plus net foreign savings (that is, the current account deficit), at the macro level there are only three ways the adjustment can play out. A higher household savings rate could be accommodated by a decline in the American current account deficit, or it could result in an increase in U.S. investment, or it could be matched by negative saving elsewhere, so that the total savings rate would not rise. There is no other way the economy can absorb a rise in the savings rate of American households.
It is important to understand the limited ways the economy can adapt to higher U.S. savings rates when grappling with the policy options.
Will Higher U.S. Household Savings Trim the U.S. Current Account Deficit?
In most other countries, a boost in household savings would almost certainly cut the current account deficit, but the United States is different. Countries that run current account surpluses must save more than they invest, and countries that run current account deficits must save less than they invest. The fact that the United States runs a current account deficit, in other words, means that U.S. savings is less than U.S. investment. Many mainstream economists would therefore conclude that any increase in U.S. household savings, by reducing the gap between investment and savings, would reduce the U.S. current account deficit.
But they are forgetting something. As the world’s borrower of last resort—the automatic recipient of the world’s excess savings—the United States operates differently than most countries. Because its deep, well-governed financial markets and its highly credible currency work automatically to absorb excess global savings from abroad, the United States is one of the few countries in the world that has no control over its domestic savings rate. The United States runs a current account deficit not because it saves too little, in other words, but rather it saves too little because it runs a current account deficit, the automatic corollary to its capital account surplus.
This dynamic has very important consequences during the coronavirus pandemic because the problem of excess savings by other countries is poised to get even worse. It is not just U.S. households who will likely respond to the economic downturn by saving more. Foreign households will also save more, and because much of this excess savings will be exported to the United States, the U.S. current account deficit is likely to rise, not fall. This means that no matter how frugal U.S. households become, higher U.S. household savings will not reduce the U.S. current account deficit: on the contrary, the country might be forced to accommodate an even higher deficit.
Will Higher American Household Savings Boost Investment?
It may seem as though higher U.S. household savings could be a shortcut to greater U.S. investment and a boon to economic growth, but that isn’t necessarily true either. If the United States were a developing economy with high investment needs constrained by scarce and expensive capital, an increase in domestic household savings rates could certainly result in higher investment. But in developed economies like that of the United States, the main constraint on business investment is expected demand, not capital scarcity. So, if anything, belt tightening among U.S. consumers could be a drag on investment, not a boost. That is why if American households decide to save more and consume less, the most likely consequence—as occurred, for example, in Germany after the 2003–2005 labor reforms that reduced wage growth and increased German savings—would be reduced business investment.
If U.S. households do reduce consumption and businesses do reduce investment—the latter condition exacerbated perhaps by a higher current account deficit—the combined effect would create a drop in total domestic demand, which would force businesses to close and lay off workers. The only way to counter this would be for Washington or local governments to engage in public sector investment in infrastructure (or incentivize the private sector to do it). Higher American household savings can only cause investment—and with it total savings—to rise, in other words, if the government responds by spending substantially more on public sector infrastructure. The private sector by itself cannot do the job.
Will Higher American Household Savings Cut Into Savings Elsewhere?
If U.S. investment doesn’t rise and if the U.S. current account deficit does not contract, total U.S. savings cannot rise. What’s more, if the current account deficit expands and business investment declines—both scenarios that are possible and even likely—total American savings actually must decline to balance lower investment and a higher current account deficit.
This reality may seem very counterintuitive, so it is worth asking: how is it possible for total U.S. savings to remain unchanged or even decline if American households react to the pandemic by increasing their savings rate? There are basically four ways this can happen:
- The drained savings of laid-off workers: Falling household consumption—perhaps exacerbated by a decline in business investment and a wider current account deficit—might cause American businesses to lay off workers. Unemployed workers produce nothing but must continue to consume, so they are forced to deplete their savings or the savings of others. This is why total U.S. household savings would be flat or even down, even though some American households will be saving a larger share of their income. The negative savings of unemployed Americans would cancel out the increased savings of Americans who were able to retain their jobs.
- Ballooning consumer debt: The Federal Reserve and U.S. banks—the former concerned about rising unemployment and the latter forced to absorb higher savings—might respond by implementing policies to encourage some American households, usually those previously unable to borrow, to take on new credit card debt. The net result would be that their higher debt (debt is negative savings) would cancel out the higher savings of other American households. As in the above case, total U.S. household savings would again be flat or perhaps even decline, even though some American households would be saving a larger share of their income.
- Boosting government spending: Washington and local governments could implement policies that create consumption on behalf of households by, for example, increasing healthcare benefits, paying for education, and otherwise strengthening the social safety net. By consuming (on behalf of households) government spending could balance out the lower household consumption caused by the pandemic. But even if that happens, higher household savings would be matched by lower government savings.
- Redistributing income: Washington could implement policies that redistribute income from wealthier Americans, who save a higher portion of their income, to poorer Americans, who save a lower portion. If such policies are enacted, U.S. household savings wouldn’t rise because higher savings among poorer households would be matched by lower savings among wealthier households.
How Will the U.S. Economy Adjust?
Total U.S. investment is by definition equal to total American savings plus net foreign savings (that is, the current account deficit), which means that there are a very limited number of ways that the U.S. economy can respond to a rise in the household savings rate. These are listed below. Some scenarios are not entirely outside the realm of possibility but seem highly unlikely.
- One possible scenario is that the U.S. current account deficit could contract, but that is very unlikely, and if the rest of the world also responds to the pandemic by saving more, it is more likely that an incoming deluge of excess foreign savings will force the U.S. current account deficit to expand even more.
- Another possibility is that U.S. businesses could increase investment, but that seems like a longshot too. Because a higher household savings rate and a constant or bigger current account surplus would reduce aggregate demand, business investment is more likely to contract.
- The Federal Reserve and American banks could try to reverse the higher household savings rate by encouraging consumer debt—usually among the poorest households—in which case rising consumer debt by some households would balance out the increase in savings by others. This approach would merely postpone, not resolve, the problem of weak demand, and it would increase the financial fragility of the U.S. economy.
- Washington and local governments could strengthen the social safety net and increase consumption on behalf of U.S. households.
- Washington could redistribute wealth downward to boost net consumption.
- Washington and local governments could also implement policies that significantly increase much-needed infrastructure investment.
- Finally, if business investment doesn’t rise, the current account deficit doesn’t decline, and government spending on infrastructure or the country’s social safety net doesn’t markedly increase, any increase in savings among some American households must result in a reduction in savings in other American households. If this balancing out isn’t caused by increasing consumer debt, it must be driven by rising unemployment.
A few other technically possible but very unlikely responses can safely be ruled out. The U.S. government is highly unlikely, for example, to allocate massive amounts of foreign aid to developing countries (thus lowering the U.S. current account deficit) in the current economic and political climate. Americans could use their savings to go on a house-buying spree, but that doesn’t seem likely either. Otherwise, the aforementioned options are the only ways the United States can adjust to higher household savings rates.
Economically speaking, the best options would be for the government to pursue efforts to build and repair American infrastructure, redistribute wealth downward, and/or strengthen the country’s social safety net. Next best would be to reduce the current account deficit by intervening to keep foreigners from dumping their excess savings in the United States. The worst options would be to reverse higher household savings by encouraging consumers to embark on a credit-fueled spending spree, and, of course, to allow unemployment to rise. Unfortunately, the country has no other practical options.
Comments(32)
Thank you for a great article, Professor Pettis!
Other countries will reduce savings. They have less cushion, on individual levels, than do US citizens. Take italy for example, starvation may be imminent as tourism dependent jobs were a large part of employment. 35%? Borrowing will increase as will debt (defacto not dejure) writeoffs.
Reducing savings is not the same thing as reducing the savings rate. If income drops, a higher savings rates may be consistent with lower total savings, but that only makes the problem described in this essay worse, not better. The key is how demand is affected.
I think it would be useful to extend this framework to the negative interest rate scenario that the markets seem to believe is at least within the realm of possibility.
the usual well argued and logical exposition to which i will add: Or, as in the 1930s, we end up with some mix of de facto/de jure trade and capital controls around continental production zones (rather than the 1930s formal imperial blocs) as states try desperately to limit unemployment and its associated fiscal stresses, and limit other countries'/blocs efforts to export their unemployment. but this, as i imagine you and klein argue in the new book, is a function of specific domestic political conflicts.
I agree that this eventually ends with capital controls. There is no reason -- except to benefit banks -- why even large economies like that of the US, let alone smaller ones like Switzerland, should be distorted by massive capital movements.
Europe is in Deflation and US entered Deflation. We shall see Deflation and Debt Deflation. We will see a Deflation Spiral Down & Deflation Collapse. Social Mood has turned Negative across the Globe.
Great piece, Michael, as usual; given the alarming breakdown in international relations, is it even possible to envision a coherent international response to the effects of the sort of seemingly intractable imbalances you have so well mapped out, especially with COVID and climate change (not to mention the usual negative externalities) set to overwhelm the developing world? If not, where do you see the next dangerous distortions developing? And, even if there is a temporary fix, whom does that end up disadvantaging: in your appealing outline, developed country (especially US, but also German, even Chinese) working class interests seem at a complete standoff with elites resisting income distribution. And, one last thing, if you would be so kind: where will tax havens and intellectual property laundering end up here? Sorry for the long set of questions, thank you so much for your great work over the years
Thanks, Larry. Those questions are too long to address here, but they are the main ones and will have eventually to be answered one way or another.
Great post as usual. I looked up Federal Reserve foreign currency liquidity swap arrangements with foreign central banks wondering how they might influence the US current account and whether or not they might add to the US deficit and potentially cause US household savings to fall or unemployment to rise. The amounts are quite large around 791,000,000,000 for terms of between 7 and 90 days. I checked the spread sheet again and it gives that total. 791 billion really is a very big number. And I guess it that number were to stay high for a quarter or two it would have a material impact. Perhaps swaps don't impact the current account but it seems to me they should. If they do then its another example of the FED along with the rest of the US financial system is prioritising the welfare of the wealthy members of the rest of the world along with the US banks and their wealthy customers over the welfare of the average US household.
One notable concept that does not appear in the piece: inflation. It would be interesting to get your thoughts on how inflation may factor into the analysis and impact the different potential outcomes. As just one example: how does the governments policies of buying trillions of financial assets (mostly loans) impact inflation generally, and/or pockets of inflation in financial assets. Surely the magnitude of the financial purchases will interact in a number of complex ways with the real data you discuss; investment, savings, GDP growth, etc. Thank you for your continued insightful analysis and congratulations on the release of your new book.
Much monetary expansion is leading to asset price inflation, and not the kind of inflation we usually mean, but for me inflation is a kind of tax, and its impact depends on the direction of the effective transfers.
Michael, thanks for a great piece. One issue: Is it really *highly unlikely* that "Americans could use their savings to go on a house-buying spree"... I'd argue the contrary, that it is a very strong possibility: 1. US Demographics mean that Millennials are starting families so there is a huge wave of demand for US housing ownership; 2. The 2005-6 housing over-build has been absorbed, so there is limited supply, especially at entry-level; 3. Interest Rates (Mortgages) should remain very low, favoring buy over rent; 4. Baby Boomer parent savings could be used for Millennial down-payments (how else will the Baby Boomers actually get any grandchildren?!), 5. The Pandemic experience gives extra incentive to own a home (It sucks to live with your parents or to rent from a mean-spirited landlord, and home owners seem better-protected politically from foreclosure). Seems to me like the US is ripe for a sustained household formation, home-building and home-purchasing boom, with all the associated consumption that creates. What am I missing?
You may be right, but it usually takes confidence and bright prospects before people buy homes, and it may be a while before these return.
I jut sold a house to a couple, both of whom work at a Fast Food restaurant, first home, I came down about 5-6% off of price, it was at cheaper end of scale for a wealthy region, where high taxes impact home values lower. Anecdotally, the real estate agent has had his best year ever. He has not slowed down since covid. There are work around's to the public agencies being close or limited in operations. The real key is affordable housing, and in many respects due to low rates and asset inflation, I have noticed these being priced out of market these last few years. I have access to the MLS, so can search across spaces in multiple states. We shall see waht happens, spending will be down, and savings up, because no one is spending money, and the ability for many to access unemployment easily, then unemployment spruced by a 600 federal add-on, makes many people doing even better than normal times. Debt jubilee is around the corner. Then direct non-debt print for infrastructure and things like semi-conductor industry
Andrew O, what are you missing? In my opinion, what Stevens shows in the first line of his response "a couple, both of whom work in the fast food industry." Income inequality, subservient service jobs that do not create real wealth but provide catering to people doing better than you are. Student Debt. American society isn't structured to allow people to work for themselves.
I wonder how infrastructure/redistribution/safety net can be the best economic options when none of those options are subject to market discipline. Now, restricting capital inflows sounds like another affront to the market, but it symmetrizes an otherwise asymmetric (and unsustainable) dynamic. I think this is the best option. Infrastructure, notwithstanding the inherent inefficiency of most such activity, is second best--it's a necessary evil. Rendering it more robust, more decentralized, to include plenty of distributed solar and SMRs, would be a worthwhile goal that matches up economic and national security priorities. Another job creating project is the repatriation of manufacturing, which will be more rapid if goosed with Federal funding. This happens to also be a national security project. This type of national industrial plan is practical, especially when paired with capital controls. Of course, it violates in a pedantic sense market discipline, but this temporary and is only necessary due to the loss of manufacturing during the system of dollar hegemony and trade deficits. I'm inclined to think that the worst option is further expansion of the social safety net, in large part because such expansions are always irreversible. Government programs, and particularly handout programs, never ever end. Let us not intensify our demosclerosis due to passing difficulties. The currently available programs will be utilized by far more people during the economic contraction, providing a time-limited boost. I would not oppose another national dividend like the $1200 recently printed, even if it were substantially more generous. But, I oppose any new handout constituencies to bribe the public and further distort the political system.
Great piece, I've learned a lot from your writing thank you. You've written in the past something along the lines that productive investment is productive because it "pays for itself". What does that mean exactly? Does it mean the government collects higher taxes in the future? Does it mean long-term GDP increases? Or something more qualitative like welfare is improved that might be more difficult to measure. In other words does "pays for itself" rely on accounting or a qualitative discussion of costs vs benefits? Spending on end of life care might be one example that highlights the difference. Or technology that decreases GDP.
It pays for itself if it boosts the value of the economy (for which GDP is often a proxy) by more than the amount of the debt. In that case the additional demand created by the spending is matched by the additional supply, so that there is no inflation.
Would it make it more likely China may be forced strategically to begin move away from its export model to a higher share on consumption, given the low marginal returns on investments? Huge near-term pain but longer term benefits?
More likely. China has been trying to do that, to "rebalance" since at least 2006. Since then the worker wage share of GDP has reduced further. To rebalance Workers wages need to grow more quickly than GDP growth, so that proportionally, more of GDP is returned to workers. Then Workers, as consumers need to spend more, that is, as workers wage share to GDP rises, Consumption to GDP need rise. Are they gonna do this, haven't so far. Michael suggests selling of gov assets, likely to be local, rather than central government and then of the government to spend on behalf of workers, this can help to rebalance by direct govenrment spending, on say the social safety net to enavle them to
Given the state of Sino-US relations and strategically speaking, do you think it possible that China will bite the bullet and move away from its export model to a more domestic-driven economy i.e. a greater share of consumption and less financial repression transfers (adding to near-term pain for the banks)?
Worker Wage Share of GDP is 50%, compared to near 70% in US. To move that 20%, it need to grow 40%, and that 40% need occur in small parts above GDP growth each year. Domestic Consumption is the HOLY GRAIL. So should China have, of course. Can it, is the quetion, ot should. The should was to work on it in 2006, the possibility is the question now. They have lowered 10% their worker to wage share to GDP since 2000 I believe, so, now they have to grow their 50% number, b 40% over annual GDP growth. The quickest way to make all that happen, is to virtually halve GDP, without losing all your employment, like the US during the Great depression. Then your worker wage share will be a higher percent of GDP, and you can have rebalanced to domestic consumption, at the cost of depression.
Paul, it's not an "export driven model" - it's an investment driven model. The people who profited the most during the period of growth, and became extremely wealthy and powerful, don't see, and cannot be made to see, doing things any other way. The wealthy and the powerful see no advantage to sharing.
Interesting how from this point of view the US is far from enjoying the exorbitant privilege of the dollar as world's main reserve currency. On the contrary, the US accepts exporting millions of jobs via current account deficit. It's very hard to imagine the Euro area sacrificing jobs in such a way!
The market response to increased savings in the face of steady income is to lower prices. Not necessarily volumes, though that too would be the case initially. What that does affect is the debt carriage of companies facing reduced revenues due to lower consumer spending that is not compensated for with the decline of input costs. Fortunately, low commodity costs have compensated for much of the revenue decrease in the "non epicenter" economy. On the debt side, companies are restructuring debt via refinancing in a bond and bank market aggressively supported by the Fed's balance sheet expansion/QE. Investment would be made when (Poutput -Pinput) X Volume change expected, is greater than the investment cost. High unemployment inducing lower pay scales and subsidized interest rates (courtesy of Fed intervention) reducing financing costs, and low commodities prices means that there is a much sharper skew towards investment than in the recent past. Second issue is that companies are refocused on resilience rather than costs and are duplicating existing capacity in China outside of it and some of that is within the US, where materials are cheaper. Third is the surge of Millennial outmigration from city centers to the suburbs/exurbs courtesy of the Work From Home movement, as household formation and births are coming for the upper half. The main US impediment to growth has been the choking point of city residential real estate costs which may be relieved by outmigration via "work from home" and distributed offices. Zillow and Redfin surveys indicate about 28% of city office workers and their employers both want WFH and intend to move out of town. Employers intend to recruit more broadly geographically as it saves them ~$40k/employee/year on average to hire outside of coastal city centers without demanding relocation. The prevailing counter-argument to WFH that office costs are only 6% of costs for service companies but labor is 60%, misses the point that the savings are on both fronts, not just the one.
It took reading your book on trade wars / class wars for me to finally understand your views and argument. If i understand our economic reality correctly, we should feel free to spend the surplus wealth accumulated by Germany and China and loaned to us at 1% rates. Indeed, i can see an argument that our long term goal should be to more or less concentrate our public and possibly some elements of our private debt in federal debt, in exchange for those surplus savings from overseas in the form of dollars.
Hi, great stuff. Only discovered you last night. I see that current account deficit has increased significantly since you wrote this. Just a quick opening question. When you refer to savings, not just in this article but throughout your work (you often speak of savings gluts) are you referring to the twin situation where consumption is reduced relative to output and at the same time deposits at banks increase? Cheers
Hi Dingo. I'm not Mike, but I agree with your assessment of a Savings Glut. I see those savings deposits at banks are better understood as expressions of indebtedness.
Great piece as usual Michael. Adding a wrinkle to this is the extension of the unemployment benefits. What happens to the savings rate if those that are unemployed continue to receive money from the government? Granted the money will be debased in this case
But they are not. Anyway, granted by whom, the common masses of influenced by the dominant ideology related to the mattes, hearing and believing, but ignorant as to mechanisms. You realize Weimar purposely debased their currency to null French reparations?
@cstevens, so you are saying the unemployment benefits have stopped? That our government had not distributed large sums of money to citizens and companies?
Comment Policy
Comments that include profanity, personal attacks, or other inappropriate material will be removed. Additionally, entries that are unsigned or contain "signatures" by someone other than the actual author will be removed. Finally, steps will be taken to block users who violate any of the posting standards, terms of use, privacy policies, or any other policies governing this site. You are fully responsible for the content that you post.