The economic impact of increased US savings
US consumers are spending less and saving more. The economic impact of that combination will depend upon how fast incomes grow.
MARCH 2009 • Charles Atkins and Susan Lund
Source: McKinsey Global Institute
Two forces that until recently turbo-charged US consumer spending—growing household debt and a falling savings rate—have gone into reverse. In late 2008, as households started reducing their indebtedness and saving more, consumption tumbled.
New research from the McKinsey Global Institute shows that the economic impact of further US consumer deleveraging will depend on income growth. Without it, each percentage point increase in the savings rate would reduce spending by more than $100 billion—a serious drag on any recovery. Relatively healthy income growth, on the other hand, would help households reduce their debt burden without trimming consumption as much.
The significance of any fall in consumption could be profound.US consumers have accounted for more than three-quarters of US GDP growth since 2000 and for more than one-third of global growth in private consumption since 1990. These trends were fueled by a surge in household debt,1 particularly after 2000 (Exhibit 1), and a decline in the personal savings rate—to a low of –0.7 percent, in 2005. From 2000 to 2007, US household debt grew as much, relative to income, as it had during the previous 25 years.
Appreciating household assets—the “wealth effect”—enabled consumers to spend and borrow more even as they saved less. The value of US household assets rose by some $27 trillion from 2000 through 2007. Rising home values, as well as stocks and other financial assets, accounted for more than two-thirds of this gain.
This dynamic sputtered to a halt when the housing bubble burst and the financial and economic crisis ensued. Falling values for homes, stocks, and other assets have battered US households: from mid-2007 through the end of 2008, their net worth fell by roughly $13 trillion. These recent losses erased all the gains in net worth, relative to disposable income, since the early 1990s (Exhibit 2). It’s not surprising that US consumer spending fell at a 4.3 percent annual rate in the fourth quarter of 2008—a major reason for the broader economic contraction.
The flip side of falling consumption is a rising personal savings rate, which reached 3.2 percent in the fourth quarter of 2008. Net new borrowing by households also has fallen sharply from its 2006 peak. In the fourth quarter of 2008, it turned negative for the first time since World War II (Exhibit 3).
Several forces underlie these shifts. Some households are responding to worries about possible unemployment or underwater mortgages by paying down debt or avoiding new debt. Others have found their credit lines shut down or can’t get new credit, because banks have tightened their lending standards.
How far these trends will go is a critical economic uncertainty in the months ahead. The economic impact of today’s deleveraging will depend on how it unfolds—through income growth, higher savings, or some combination of the two.
If incomes stagnated, for example, households could deleverage only by saving more. Every percentage point reduction in the debt-to-income ratio would require nearly a one percentage point increase in the savings rate. The US personal savings rate reached 5 percent in January, 2009. If this level prevailed and incomes didn’t grow, this would reduce the household debt-to-income ratio by five percentage points—which still wouldn’t be enough to restore the levels of indebtedness prevailing in 2000, before borrowing started to accelerate.
But if incomes rose, households could both reduce their debt burden significantly over time and continue to consume. If US incomes grew by 2 percent a year, for instance, households could reduce their debt-to-income ratio by as much as they would in the scenario above—but with a personal savings rate of only 2.3 percent.
These different scenarios have serious implications for the US and global economies because, holding incomes constant, each percentage point increase in the savings rate translates into roughly $100 billion less in consumer spending¬. A 5 percent savings rate would mean $530 billion less in spending each year if US incomes fail to rise; if they rose by 2 percent a year, a 2.3 percent savings rate would mean $250 billion less spending, all else being equal.
In short, the importance of income growth is difficult to overstate. With it, households can simultaneously reduce their debt burden, rebuild savings, and boost consumption. But without significant income gains, deleveraging could undermine consumption and the global economy for years to come. One implication: policy choices that favor productivity and employment growth—critical determinants of income growth—will make deleveraging less painful. Efficiency breakthroughs in sectors, such as health care and government, that employ large numbers of people—but that have not enjoyed productivity revolutions similar to those experienced in industries like retailing and wholesaling—would make a dramatic difference.
About the Authors
Charles Atkins is a consultant in McKinsey’s San Francisco office, and Susan Lund is a consultant in the Washington, DC, office.
Francisco Vergara
President
ADEP
Paris, France
In your excellent article “The economic impact of increased US savings,” you conclude by writing that “health care and government … have not enjoyed productivity revolutions similar to those experienced in industries like retailing and wholesaling.”
Are you sure? The same computers have been put in everywhere. This could just be a statistical illusion due to the different ways productivity is calculated in different branches. It is very difficult to define exactly what health care and government ‘produce’, so it is very difficult to measure how much more of it per hour (or per worker) is being produced. So an assumption is made. As the OECD writes, “the methods used by most countries to estimate value added in government services assume that labour productivity growth is zero,” OECD Factbook 2008, p. 264.
Supposing it isn’t a statistical illusion, and that productivity in health care and government started to grow faster? Would employment grow or decline in these branches?
This productivity thing is trickier than you think.
Jim Hansen
Finance Director
Henry V Events
Oregon, United States
Is there any research that tells us the composition of the consumer spending fueled by the massive increase in household debt? Knowing what percentage of the new debt went to purchasing automobiles, kid’s college educations, health-crisis care, and 2nd homes (future retirement homes?) versus vacations, clothing, consumer electronics, and entertainment, would help us gauge how consumer markets might rebound over time.
Mike Woods
President
Woods Consulting, Inc
Santa Monica, CA USA.
A very good article, but it would be helpful to separate the effects of mortgage equity withdrawals out from other credit sources. This would underline the significant impact of the housing bubble and provide insight to why an implosion in one sector has cascaded to many others.
Also of interest is the impact that the bailout and stimulus actions by the Federal Reserve and Treasury Departments will have on this picture. By generating inflation and by driving down the value of the dollar, these actions will add more support to prop up the debt-to-income ratio.
Although this is an academic paper with limited scope, the next questions are the most interesting. What will be the impact to our society from these forces?
Is it right to encourage additional spending when individuals are so overextended? How do we justify a future in which people’s debt load continues at greater than 100 percent of income? In essence, this creates additional risk for individuals in the hope of creating broader stability.
At what point is economic contraction necessary in order to deleverage society to levels that are sustainable into the future?
And how would ‘efficiency breakthroughs’ in government-controlled segments, such as government and healthcare, lead to a dramatic difference? Haven’t we already learned from this crisis that excessive promises to unions in the forms of employment- and wage-stability and guaranteed pensions create liabilities out of balance of industry incomes, leading to massive deficits in government spending and the zombie state of the auto makers?
Robert King
Economist
The Jerome Levy Forecasting Center
Mt Kisco, NY
In theory, of course, we could experience a rising personal saving rate in aggregate with no decline at all in household debt; for instance, if the increase in saving were entirely concentrated among households with no outstanding debt. In practice, a considerable amount of increased saving will likely go toward paying down debt, but the relationship between rises in the saving rate and declines in the household debt-to-income ratio is probably not one-to-one. More importantly, income and employment declines themselves—especially severe ones like those of the current period—exert significant downward pressure on the personal saving rate.
John Calia
Managing Partner
Tatum LLC
Fort Lauderdale, FL
The open question, of course, is what policies would improve productivity and employment growth? Better yet, would the Obama administration support them? I have long supported the eliminaton of the corporate income tax as a way of making employment costs cheaper in the US. Yet, I don’t think it’s politically viable.
Posted by : hsr ( April 2, 2009 )
|
BIZ_TALKS SITES