In the past month, there have been two articles ("American Savings Understated by Most Conventional Measure," Marc Chandler, Financial Times. March 8, 2005 and "Running on Empty?," David Malpass, The Wall Street Journal, March 28, 2005) in which the authors argue that household saving is being underestimated. My interpretation of their conclusion is, to quote Bob Marley, "Don't worry 'bout a thing, 'cause every little thing gonna be all right." I respectfully disagree with their thesis. Rather, I will argue that this is one of those rare cases when the conventional wisdom is correct - that is, households are saving very little to the detriment of their future standard of living.
One of Chandler's and Malpass' arguments is that personal disposable income has been underestimated in recent years due to the behavior of private pension funds. For example, the corporate contributions to employee pension funds are counted as income to the employees in the period when the contributions are made. But during the NASDAQ bubble years, corporate contributions were curtailed because some defined benefit pension plans appeared to be overfunded due to extraordinary holding period gains of corporate equity holdings in these plans.
Let's not concern ourselves with these national income accounting nuances, but get to the heart of the matter -- in recent years, has household borrowing, a flow concept, risen relative to household spending, also a flow concept? The answer is unequivocally "yes." Chart 1 shows the dollar-value change in total household liabilities (from Federal Reserve flow-of-funds data) as a percent of the dollar-value of total household spending. Total household spending is defined as the sum of the GDP line items Personal Consumption Expenditures and Private Residential Investment Expenditures. In 2004, households' total borrowing represented 12.5% of their total spending - the highest percentage since the 1952 start of the series. If households' incomes are so underestimated, why are they borrowing so much relative to their spending?
Chart 1
Households: Change in Liabilities ($) / Total Spending* ($)
%
* Sum of Personal Consumption and Private Residential Investment Expenditures
Another way to get around the alleged underestimated after-tax income issue is to look at households' net acquisition of financial assets - stocks, bond, deposits, pension fund reserves, etc. compared to their net acquisition of liabilities - borrowing. Households' net acquisition of financial assets represents net purchases or net additions to their previous holdings of financial assets. So, their net acquisition of financial assets does not include holding period gains on previously-acquired assets. How does one go about acquiring financial assets other than "marrying up" or inheriting them? By spending less on goods, services and tangible assets than your cash income. By definition, whatever is left over must represent your net acquisition of financial assets, even if those assets are only bank deposits. What if you spend more than your cash income? Can you still acquire financial assets? Yes, if you borrow. Let's put this in equation form:
(1) Net Acquisition of Financial Assets = Cash Income - Cash Spending + Borrowing
Now, let's re-arrange some terms:
(2) Cash Income - Cash Spending = Net Acquisition of Financial Assets - Borrowing
From Equation (2) (really, an identity), we can see that if borrowing is greater than the net acquisition of financial assets, that is, the right-hand side of the identity is negative, then the left-hand side of the identity must be negative, too. Another way of saying this is that if our borrowing exceeds our net acquisition of financial assets, then we are spending more cash on goods, services and tangible assets than we are taking in. Some might say that we are dissaving.
Now, why do I take you (those few who still are reading) through this tortuous exercise? Because our friends at the Fed provide the relevant data in their flow-of-funds accounts to make the necessary calculations to determine whether households are saving or dissaving. These data are presented in Charts 2 and 3. Plotted in Chart 2 are households' net acquisition of financial assets and households' net acquisition of liabilities, or household borrowing. Chart 3 plots the right-hand side of Equation (2). As can be seen in Charts 2 and 3, starting in 1999, household borrowing started to exceed households' net acquisition of financial assets. In other words, starting in 1999 and continuing through 2004, households' cash outlays on goods, services and tangible assets have exceeded their cash incomes. From 1952, the beginning of these data series, through 1998, this phenomenon of households spending more than they were taking in had never occurred. If spending more than you take in is evidence of a strong propensity to save, then perhaps George Orwell's Ministry of Truth has actually come into existence and is being run by Chapman and Malpass.
Chart 2
Source: Federal Reserve Board /Haver Analytics
Chart 3
Source: Federal Reserve Board /Haver Analytics
Now, Malpass might counter that household spending on SUVs and McMansions is a form of saving because these tangible assets yield transportation and shelter services currently and into the future. But the interesting economic question is whether this "saving" will enable households to earn higher incomes in the future so that they can service their increased indebtedness and still enjoy a rising standard of living? That is, will "investing" in SUVs and McMansions enhance labor productivity? Are those expensive leather seats in SUVs going to make households more productive? Is that expensive 4-wheel-drive feature on the SUV going to get Floridians, who encounter neither snow nor steep inclines, from Point A to Point B faster? Are those granite kitchen countertops going to make households more productive? Another way to think about this is to consider the hole-drilling corporation that uses the proceeds from its stock and bond sales to "invest" in fancy office furniture and other frills instead of state-of-the-art drill presses. Will fancy office furniture enable the corporation to earn higher profits in the future? If not, how will it make future payments of interest on its debt and dividends on its stock?
I want to once again address this issue of holding period gains as saving. (For a fuller discussion of this issue, see "Wealth Illusion" http://www.northerntrust.com/library/econ_research/weekly/us/pc102204.pdf.) There is no doubt that holding period gains increase one's net worth today. But are these holding period gains saving in the economic sense? That is do these holding period gains represent the building up of capital - both human and inanimate - that will enable goods and services to be produced at a faster rate in the future? Does the increase in the price of an existing house represent an increase in the amount of real shelter services provided by that house? Evidently not. Plotted in Chart 4 is the real implicit rent on owner-occupied housing, a measure of the real shelter services provided by owner-occupied housing, as a percent of the market value of that housing. Also plotted in Chart 4 is the dollar value of holding-period gains on households' real estate assets. As can be seen, despite the fact that the market value of housing has risen sharply in recent years, its "yield" of real housing services has continued to trend lower.
Chart 4
Essentially there are two ways for households to increase there net worth - spend less than they earn or be fortunate enough to enjoy holding period gains on assets already owned. As Chart 5 shows, starting in the second half of the 1990s, holding period gains have contributed more to household net worth changes than usual. It could be argued that holding period gains could actually retard future economic growth prospects. How? A rapid increase in household net worth due to holding period gains could induce households to spend more of their current income rather than investing in businesses through the purchases of corporate stocks and bonds. Businesses typically spend the proceeds from stock and bond sales on capital equipment, which, in turn, enables the economy to grow faster in the future. So, increases in household net worth emanating from holding period gains could lead to less saving in an economic sense and, therefore, slower future economic growth prospects. By the way, it is not difficult for a government to increase household net worth through holding period gains. All it has to do is induce the central bank to "print" more money. The printing of more fiat money will drive up prices, including the prices of assets owned by households.
Chart 5
Lastly, I want to leave you with one chart that sums up the profligacy of American households. Chart 6 shows total household spending as a percent of GDP (all in nominal terms). This percentage started drifting up in the second half of the 1980s but then spiked up in the late 1990s. In 2004, total household spending represented a record 75.8% of GDP. Today's "partying" in terms of a disproportionate share of national output being "consumed" by the household sector is a recipe for a "hangover" tomorrow.
Chart 6
Household Total Spending* ($) / Nominal GDP ($)
%
* Sum of Personal Consumption and Private Residential Investment Expenditures
