I love the Fed's quarterly flow-of-funds report. It usually is the mother lode of enlightening economic nuggets of information. And the Fed's latest release on December 7 of third-quarter data was rich with these nuggets. For starters, the decline in U.S. bond yields in recent months is less of a mystery when you take into consideration the sharp slowdown in the rate of domestic nonfinancial borrowing. Chart 1 illustrates the point. Relative to nominal GDP, nonfinancial domestic borrowing (i.e., the annualized dollar change in debt outstanding) peaked at 19.7% in Q4:2005, moving down to 13.9% in Q3:2006 - the lowest percentage since Q4:2003.
Chart 1
Although not the only nonfinancial sector accounting for this slowdown in borrowing, the household sector was the principal one. Chart 2 shows that after hitting a post-WWII high of 14.6% in Q3:2005, household borrowing relative to disposable personal income (DPI) dropped to 8.8% in Q3:2006 - the lowest since 7.6% in Q3:2001, when the economy was in a recession. Notice in Chart 2 that precipitous declines in this percentage tend to be followed by the onset of economic recessions (indicated by the shaded areas in the chart).
Chart 2
Just to demonstrate how precipitous the current fall off in household borrowing has been, I had Haver Analytics calculate the year-over-year change in household borrowing relative to DPI. This is shown in Chart 3. Wow! The percentage is down from year-ago by 5.8 points - the largest decline since Q2:1980, when President Carter urged us to don sweaters and tear up our credit cards.
Chart 3
But in the current situation, households have not been cutting up their credit cards but rather sharply scaling back the growth in their mortgage credit as the housing market recedes. This is shown in Chart 4. The most recent year-over-year decline in household mortgage borrowing as a percent of DPI is unprecedented in the post-WWII period.
Chart 4
At the same time that households are slowing down there borrowing relative to their after-tax income, corporations also have slowed their liabilities issuance relative to their after-tax income. This is shown in Chart 5.
Chart 5
Unlike households, corporations can issue equity. So, the liabilities of corporations consist primarily of credit market debt instruments and equities. Household liabilities consist mainly of credit market debt instruments. As I have discussed in previous commentaries, corporations have been in engaged in the "retirement" of massive amounts of their equities, either through stock buybacks or through mergers. Chart 6 shows that in recent quarters, corporations have been retiring near-record amounts of their equity relative to their after-tax income. The record relative retirement occurred in the late 1990s. Hmm. Let's see. The stock market was going up nicely in the late 1990s and it has been going up not as nicely, but nicely nevertheless of late. And in both periods, corporations were sharply retiring equity. Econ 101 would suggest that this might put a bid in the stock market.
Chart 6
But there is at least one major difference between the current environment and the late 1990s. Today, corporate credit market borrowing relative to after-tax profits is only about half what it was back in the late 1990s. Back in the late 1990s, as I pointed out back in the late 1990s, corporations were borrowing in the credit markets to fund their equity retirement. That is, corporations were levering themselves. Today, corporations are using more of their profits than borrowed funds to retire equity.
At the same time that the demand for funds on the part of the U.S. domestic nonfinancial sector has been slowing, the supply of funds from foreign investors keeps growing faster than U.S. nominal GDP. This is illustrated in Chart 7. Net foreign financial investment is the increase in foreign investors' net acquisition of U.S. dollar-denominated securities - stocks and bonds - minus the net increase in dollar-denominated liabilities incurred by foreign entities. Net foreign financial investment in the U.S. has increased to over 6% of nominal GDP.
Chart 7
In sum, in the past few quarters, we have seen a sharp slowdown in household borrowing and a sharp increase in corporate equity retirement funded out of current profits. At the same time, we have seen continued strong growth in funds being advanced to the U.S. This helps explain both the rallies in the bond market and the stock market. With regard to the bond market, this also helps explain the inversion of the yield curve. The supply of credit from abroad has continued to grow rapidly as the growth in the U.S. demand for credit has slowed sharply. Econ 101 posits that when supply grows relative to demand, the price tends to fall. The price of credit is the interest rate. The relative excess supply of credit has put downward pressure on U.S. bond yields. But the Fed, by targeting the overnight cost of funds, the federal funds rate, has prevented shorter-maturity interest rates from falling in tandem with bond yields. The way the Fed prevents the funds rate from falling is by supplying fewer reserves than would otherwise be the case if short-maturity interest rates were allowed to adjust to non-Fed supply-demand credit conditions. A falling bond yield in the context of an inverted yield curve is not a sign of "easier financial conditions." Just the opposite. This is why the spread between the Treasury bond yield and the fed funds rate is a leading indicator. A widening spread is an indication of an easier monetary policy; a narrowing spread is an indicator of a tighter monetary policy. Almost every time, including the current environment, this spread narrows to the point of turning negative, with a lag, real economic growth slows. In case you had not noticed, U.S. real economic growth has slowed.
Despite the fact that household mortgage borrowing has slowed in recent quarters, the leverage in owner-occupied residential real estate reached a record high 46.4% in Q3:2006, as shown in Chart 8. If mortgage borrowing slowed, why the increase in leverage? Because, as shown in Chart 9, there has been a sharp slowdown in the growth of the total market value of residential real estate. With a still -sizeable excess inventory of homes for sale, continued weak growth, perhaps even a contraction, in the market value of residential real estate could reasonably be expected in 2007.
Chart 8
Chart 9
With the sharp slowdown in the growth of housing values, it is quite natural that there also would be a sharp slowdown in the growth of homeowners' equity. This, combined with higher adjustable rate mortgage financing rates, has resulted in a sharp slowing in mortgage equity withdrawal (MEW). As shown in Chart 10, MEW peaked at an annual rate of about $730 billion, or 8.1% of DPI in Q3:2005, slowing to an annual rate of only $214 billion in Q3:2006. Along with corporate stock retirement, MEW has been an important source of funding for household deficit spending in recent years. Therefore, this slowdown in MEW would be expected to slow the growth in household spending, which, as shown in Chart 11, has begun. On a year-over-year basis, growth in the sum of real personal consumption and residential investment expenditures has slowed to 2.0% in Q3:2006, the slowest growth since the past recession.
Chart 10
Chart 11
Household liquidity fell to a post-WWII low in Q3:2006 (see Chart 12). I am using as a measure of liquidity household deposits and money market mutual funds as a percent of total household liabilities. Some might respond that with all the different sources of credit available to households today, they do not need to hold as large a ratio of liquid assets as in the past. To this I would respond with three counter-arguments. Firstly, households already have borrowed so much that their leverage ratio is at a post-WWII high (see Chart 13). Secondly, households have already borrowed so much that their debt service burden is at a 25-year record high (see Chart 14). And thirdly, residential real estate, which accounts for 30.5% of the total market value of household assets (see Chart 15), is the single largest asset in households' portfolios compared with deposits, credit market instruments, corporate equities (about 44% of which are held on their behalf in pension funds and insurance companies) and other tangible assets. Of these other asset categories, residential real estate probably is the least liquid, aside from used refrigerators (other tangible assets). In sum, households have never been as highly levered as they are now or as illiquid as they are now, and their single largest asset is in danger of actually falling in value. If the Fed had to resume raising interest rates in this environment, it would be "Katy, bar the door" for household finances!
Chart 12
Chart 13
Chart 14
Chart 15
*Paul Kasriel is the recipient of the 2006 Lawrence R. Klein Award for Blue Chip Forecasting Accuracy