This week's "Outside the Box" is by my friends and the always out of the box thinkers at Hoisington Investment Management. In their 3rd quarter market commentary, Van Hoisington and Dr. Lacy Hunt weigh in on a myriad of topics ranging from the data affecting interest rates to housing to the outlook for the dollar. They go on to discuss a correlation between the yield curve and the Leading Economic Index (LEI) that has produced a very accurate track record of predicting recessionary environments.
Based in Austin, Texas, Hoisington Investment Management Company is led 2 economists, Van Hoisington and Dr. Lacy Hunt. They specialize in management of fixed income portfolios for large institutional clients by setting long-term investment strategies based on economic analysis.
While the markets seem stuck in the waters of uncertainty, it is more important than ever to continually focus on prudent principles and independent thought to yield intelligent investments.
John Mauldin, Editor
Outside the Box
Quarterly Review and Outlook
Third Quarter 2006
THIRD QUARTER RALLY
The U.S. Treasury bond market rallied sharply in the third quarter. Two, five and ten year notes returned 2%, 3.4%, and 5% respectively, while the longer-dated thirty year bond rose more than 8% for the quarter. This rally was generated by weaker economic data while the Fed funds rate remained at 5 1/4%. The equity market, as measured by the S&P, also moved upward registering a 5.7% rate of return for the quarter.
Presently, the five, ten, and thirty year interest rate levels, standing at 4.6%, 4.7% and 4.8% respectively, are noticeably below the 5 1/4% funds rate. This yield curve inversion, while unusual, is not unprecedented. Similar inversions occurred prior to six post WWII recessions. In the inversions prior to the 1990 and 2000 recessions, the yield curve became more inverted until the Federal Reserve lowered the Fed funds target rate. A similar outcome should prevail in this business cycle, with the inversion process only halting when the Fed reduces rates.
THE "NO OPINION" INDICATORS
Judgments regarding the course of future economic activity are as plentiful as pages in a newspaper. Each forecast contains assumptions about future economic developments. As a result, current opinions range between a Fed pause and subsequent tightening to an early reduction in the Fed funds rate. Interestingly a body of statistical indicators exist that carry no opinion, but which have proven over an extended period of time to do a superior job of forecasting the level of economic activity. This statistical series, compiled by the Conference Board since 1960, and prior to that by the U.S. Department of Commerce, is known as the Leading Economic Index (LEI). The LEI, comprised of ten different economic indicators, has declined on a six month basis thirteen times since 1952. Nine recessions and four slowdowns followed each of these contractions. This 70% batting average is measurably better than the record of most economic pundits.
One of the LEI components--the spread between the ten year Treasury note yield and the Fed funds rate-- has, by itself, an outstanding record of signaling recessionary periods since the late 1960s.
It is possible to improve the statistical probabilities of these two prescient indicators, by combining them into a third leading barometer. Since the end of WWII, a simultaneous yield curve inversion and a six month decline in the LEI have occurred only seven times. With the exception of an inversion forty years ago, this combination has been 100% in calling recessions The average lead time between the meeting of these two conditions has been about nine months (but with significant variations from the average). The most recent episode occurred in June of this year, suggesting that the economy should register at least below trend growth rates in 2007. All "opinions" aside, cyclical financial history points to disinflationary, sub-par growth rates over the next several quarters. Augmenting this statistical view is the present collapse of various housing indicators.
The housing sector has recently recorded the greatest boom in U.S. history. Speculation and a plethora of new financing alternatives have driven home prices to extreme levels. A significant increase in the Fed funds rate to 5 1/4% over the past couple of years has begun to impact this heretofore strongest sector of the U.S. economy. In fact, it is safe to say that housing is already in a recession. The National Association of Home Builders Market Index has fallen to a sixteen year low.
New home sales are off 17% from a year ago; resales are down 13%, and importantly, prices in both categories are slightly negative on a year over year basis. Housing starts have slipped 20% below year ago levels. While ostensibly housing only represents about 6% of GDP, the multiplier impact on lumber, copper, steel, cement, household furnishings, and employment is significant. In light of the fact that housing related employment has contributed roughly 1/3 of the employment growth of this entire expansion, its reversal will significantly influence future growth rates. Presently, construction employment is still rising modestly, so it is clear that the U.S. economy has yet to feel the negative impact of the housing reversal, insuring disappointing growth in 2007. Despite the prospect for below trend growth over the next several quarters, and the subsequent creation of more excess capacity and looser labor markets, long dated bonds are viewed as risky due to the large trade deficit and its imagined impact on interest rates.
THE TRADE DEFICIT
One of the most persistent worries concerning the bond market is the record trade deficit. As the logic goes, the deficit is so unmanageable that at any point foreign investors (who have taken U. S. paper in exchange for our "excessive" buying of their goods) may dump their dollars, forcing the dollar downward and the U.S. inflation rate upward. Our research indicates that these concerns are entirely misplaced.
First, attempting to find a correlation between the U.S. dollar and the trade deficit fails. Since 1974 the trade weighted dollar and the trade deficit register a correlation coefficient of 0. The "t" stat is a statistically insignificant .001, and carries the wrong sign. The obvious lack of correlation is confirmed by a graph of the two variables.
For example, from 1995 to 2002, the trade deficit deteriorated by 442% while the dollar gained 28%. In other words, there has never been a correlation between the U.S. dollar and the trade deficit. Assuming dollar weakness just based on trade is an unproven and highly suspect conclusion.
Second, the trade deficit and its associated internal capital flows have allowed foreigners to amass a total of $13.6 trillion in claims on U.S. assets. From the U.S. position, this could be seen as $13.6 trillion in debt or, conversely, from the foreigner's perspective they hold $13.6 trillion of U. S. assets. This could be of great concern since it does represent 106% of one year's total U.S. economic growth. However, this statistic ignores the important fact that U.S. investors hold $11.1 trillion in foreign assets. In other words, the world is in debt to us by $11.1 trillion. Of significance, the difference is $2.5 trillion, only 20% of one year's income or GDP for the United States. This ratio is about the same as the average homeowner pays for housing expenses--about 20% of income. Two Nobel Laureates--Milton Friedman and Edward Prescott--shed significant light on this complex problem. They maintain that any impact on the dollar comes from the difference in what is earned on the $11.1 trillion that we own versus what foreigners earn on their $13.5 trillion.
In Chart 4 we derived the yield on those various assets. As evident in the graph, historically U.S. investors have achieved a higher yield than their foreign counterparts. And, this differential is much more closely correlated with movements in the dollar.
The great bulk of U.S. asset holdings are in the form of direct investments in plant equipment and companies. On these assets we enjoy a higher income and a gain from the rise of their value. The rest of the world, however, invests heavily in U.S. securities that provide lower yields and less potential for capital appreciation. The most recent statistics reveal that, of the $13.6 trillion, the rest of the world held $6.4 trillion in long term securities, with $2.1 trillion in equities and $4.2 trillion in debt.
As the table indicates, the debt instruments are highly concentrated in the short end of the market. With the recent rise in the Fed funds rate, yields on short dated instruments have risen, and therefore essentially evened the yield differential in this calendar year. As short term U.S. rates fall away, the spread will move back in favor of U.S. asset holders. Therefore, concerns regarding the dollar and its impact on U.S. interest rates are fundamentally unfounded if they are based on the trade deficit or massive foreign holdings of U.S. assets.
Recent revisions in the payroll series for the year ending fiscal March 2006 revealed a much stronger recovery and level of economic activity for that period than was previously known. Further, the very modest, 103,000 job gains in private payroll employment over the past six months is called into question as the Department of Labor could also be understating present job growth. Thus, economic circumstances could be understated. This would imply that the Federal Reserve's pause is just that, and a further tightening of monetary policy would be required over the next several quarters to slow economic activity below potential. Of course this would cause the entire yield curve to move upward.
Mitigating that bearish conclusion are the following facts: a) significant restraint has already been applied to the system, as evidenced by the 5% decline in total reserves over the past twelve months, and the slow 4.7% M2 growth rate for the past year; b) housing, as previously noted, is declining sharply and new regulations setting higher standards for lending, which were just issued, will create continued downward pressure on that sector; c) fiscal stimulus has been significantly reduced this year as the deficit totaled about $250 billion compared with $318 billion and $411 billion in the two previous fiscal years; d) the leading economic indicators of world economic activity, as well as the U.S., are pointed toward much slower growth; e) commodity prices have fallen roughly 20% from their peaks, implying softer economic activity and (f) real GDP has eased to the 2 1/2% range, well below the 3.5% growth rates of the past two years.
On balance, we believe the lagged impact of Federal Reserve policy actions and mounting weakness in specific sectors will cumulate toward much slower growth in 2007. Interest rates should reflect that slower growth scenario.
Your still amused by the employment data analyst,