Insight Line—February 19, 2007

In 2006, American workers filed 16.2 million new unemployment insurance claims. But the total unemployed was more likely higher than that, as many workers who lost jobs did not qualify for insurance benefits and therefore would not appear in the weekly statistics. At first glance, the number suggests that slightly more than 10 percent of the nation’s labor force became unemployed last year. That is, if you choose to view the numbers out of context. The real story is that the American economy created 19.3 million jobs—or 3.14 million net new jobs—over the course of the year, sending total employment as measured by the household survey to 152.8 million.

But how do we decide whether the hard data is indeed reflecting the whole story? This consideration begs an oft-heard question of late: how accurate is the monthly Employment Situation report?

Well, according to research from the Federal Reserve Board (Fed) staff, not very, it seems.

Lest you conclude this is a newly minted piece of research developed in response to the rather high profile adjustments that the Bureau of Labor Statistics (BLS) has had to make over the course of the past few months to its monthly reports, it is not. The report, entitled “Employer-to-Employer Flows in the U.S. Labor Market: The Complete Picture of Gross Worker Flows,”¹ first appeared in May 2004.

Using available data adjustments to the BLS Current Population Survey, known as the household survey, the Fed’s staff in Washington, D.C. offered up a compelling challenge to Wall Street’s preferential treatment of the payroll data portion of the report. Simply put, what the researchers posited was that in our predominantly service-based economy, working Americans’ occupational categories change faster than the federal government’s BLS can track them. In fact, more than 6 percent of the nation’s labor force— that’s 9 million workers—finds a new job, becomes unemployed, or withdraws from the labor force each month. In essence, the great job churn is always at work.

Few people other than the experts know that the BLS, in its efforts to depict and analyze employment trends in this country, collects employment data from two different survey sources: the household survey and the payroll survey.

The household survey tries to establish—within a statistically significant confidence level—the size of the nation’s labor force, as well as the number of people employed. The numerical difference between the two, translated into a percentage rate of the total labor force, results in the widely referenced unemployment rate.

On the other hand, the payroll survey collects data directly from employers’ payrolls. Focusing on job sites, the monthly results ostensibly convey the number of workers added or subtracted by employers on a month-over-month basis.

For years, economists questioned the usefulness of the household data versus the payroll data in discerning trends in the nation’s labor force. They asked, is the household survey, which is based on responses from 60,000 households, as accurate a snapshot of the national employment situation as the payroll survey, which is based on responses from 160,000 businesses and government agencies covering 400,000 establishments?

But now, after a string of widely discussed revisions to the monthly payroll data, all of that may be changing.

You see, the Fed research from 2004—which suggests the existence of a very dynamic labor force “on average 2.6 percent of employed workers leave one employer for another each month”²—clearly depicts the formidable challenge facing the selection process used in the payroll survey. After all, if close to 4 million employed Americans are changing employers each month, how can the BLS, or investors for that matter, be certain that its payroll survey accurately reflects where the jobs are and where they may be?

To be sure, the BLS’s monthly Employment Situation report remains one of the most anticipated and analyzed economic reports on the part of investors, economists, politicians and policymakers. However, as I see it, those emphasizing the results of the payroll survey at the expense of the household survey—or vice versa—run the risk of ignoring the dynamic nature of the nation’s labor force.

Fortunately, as I see it, the staff of the Fed recognized this years ago.

¹ “Employer-to-Employer Flows in the U.S. Labor Market: The Complete Picture of Gross Worker Flows,” by Bruce Fallick and Charles A. Fleischman, Finance and Economics Discussion Series, Board of Governors of the Federal Reserve System, May 2004.

² Ibid, page 2.

Factors Driving the Economy

Economic Acceleration

Economic Deceleration

Latest Available Information
(as of 2/16/07)

Employment
^{(Source: Bureau of Labor Statistics)}

Up

Down

Headline employment gains in February 2007’s data were of little value. Owing to the yearly benchmark adjustment, the past 12 months’ data were markedly altered. That said, the string of record levels in both measures of employment, household and payroll, remained intact. Total employment, as measured by the household survey, clocked in at 145.9 million, while the payroll data recorded 137.3 million workers. The nation’s unemployment rate inched up 4.6 percent.
Overall, February’s data continues to suggest that economic growth is not placing undue demand on the available pool of workers.
Please circle Up on the table.

Personal Income
^{(Source: Bureau of Economic Analysis)}

Up

Down

Personal Income increased 0.5 percent in December 2006 and 6.4 percent for the year. When adjusted for price changes, real Disposable Income increased 2.7 percent, or more than twice the rate recorded in 2005. In total the data suggests higher energy costs and a weakened housing market are not presenting consumers with formidable challenges in the opening months of 2007.
Please circle Up on the table.

Retail Sales
^{(Source: Department of Commerce, U.S. Census Bureau)}

Up

Down

Advance Retail Sales for January 2007, were less than forecast, recording no change on a seasonally adjusted basis from December 2006. Confirming my earlier forecasts the year-over-year growth rate slipped to 2.3 percent from 5.7 percent. This is the lowest yo- y growth rate since the first quarter of 2003. This data release calls into question an outsized contribution to first quarter’s 2007 GDP growth. Though this drop off is not alarming due to big seasonal swings at this time of year I continue to suggest circling down on the table is the appropriate choice.
Please circle Down on the table.

Durable Goods
^{(Source: Department of Commerce, U.S. Census Bureau)}

Up

Down

According to the most recently available data, the nation’s manufacturing activity surprised to the upside. Total factory orders rose a stronger than anticipated 2.4 percent; however durable goods orders in previous months were not as large as initially reported. The report stands in contrast to the monthly manufacturing surveys conducted by various other agencies. Therefore, while the forward order books remain very promising, accumulated inventory continues at levels higher than what businesses consider ideal. This sets up a tug-of-war between production and sales that will not be easily resolved. As such, I am choosing to keep the suggested action in the acceleration column.
Please circle Up on the table.

Inflation
^{(Source: Bureau of Labor Statistics)}

Low

High

Annualized inflationary pressures, using the Federal Reserve’s preferred measure of inflation, the Personal Consumption Expenditure excluding energy and food (PCE), remained at 2.2 percent in the December 2006 data. This year-over-year measure remains above the Federal Reserve’s stated acceptable levels. As such, recent Federal Reserve Open Market Committee (FOMC) members’ public comments, irrespective of the current policy pause, suggest continued concern over inflationary pressures remains amongst board members.
Please note, though inflation, as measured in a historical context, remains low, the suggested course of action on the table is to circle high, reflecting the publicly stated concerns of the FOMC.

Government Spending
^{(Source: Congressional Budget Office, U.S. Treasury)}

$In

$Out

The fiscal 2007 budget numbers continue to defy some dire forecasts. As of January 2007, four months into the fiscal year, the year-over-year numbers detail a $56 billion improvement. On a 12-month basis, the deficit as a percentage of gross domestic product (GDP) now rests at a multi-year low of 1.4 percent. This change of events has prompted Treasury officials to discuss curtailing some anticipated borrowings over the course of the fiscal year. That said, the longevity of this pattern will not reveal its true nature until after the April personal income tax date.
Please circle $In on the table.

Monetary Policy
^{(Source: Board of Governors, the Federal Reserve System)}

$In

$Out

The FOMC’s target for the federal funds rate remains unchanged since June 29, 2006 at 5.25 percent. Public comments from members of the FOMC leave little doubt of lingering concerns over the expected course of future inflation. In the Monetary Policy Report to Congress FOMC Chairman Ben S. Bernanke suggested that the core inflation rate remains somewhat elevated. As such, barring any dramatic economic developments, the first chance for a rate cut now moves to the June FOMC meeting.
Please circle $Out on the table.

Yield Curve
^{(Source: Bloomberg, LP)}

Positive

Negative

The difference between the three-month Treasury bill yield and the 10-year Treasury note yield was -32 basis points on February 16, 2007.
Though the shape of the curve remains inverted, it has yet to approach the point where historical analysis suggests heightened concern over the future course of economic activity.
Please circle Negative on the table.

Gross Domestic Product
^{(Source: Bureau of Economic Analysis)}

Above

Below

The advance report of the fourth quarter 2006 GDP places the annualized growth rate at a stronger than expected 3.5 percent. The data clearly revealed positive contribution from consumer spending and a worrisome subtraction from residential investment and inventory management. For all of 2006, GDP as reported expanded at a real rate of 3.4 percent versus the 3.2 percent recorded in 2005. However, upcoming revisions to the data may reduce growth somewhat lower than initially reported.
Please circle Below.