US Perspectives: What's Not Working in the Jobs Report

The unemployment rate dropped by a stunning 0.4 percentage points for the second month running. Only about 0.1 percentage points of the drop was related to a smaller labor force participation rate. Consumers might feel just a little bit better about their employment situation with the rate now dropping back to 9%. Maybe that's why recent spending reports--such as the International Council of Shopping Centers January report and the government's Personal Income and Outlays report for December--have all looked so robust.

All in all, I am feeling better about the economy last week, but I will keep the yellow flag out for now until I get a more conclusive employment report and more inflation reports. While we wait for more definitive job data, the payroll tax should keep the economy humming for a least a few more months.

Despite Today's Report, Employment's on the Mend; the Bond Market Gets This

As confusing as the government report is, when I dig into it and combine it with other job reports, it is very clear to me the employment situation is improving, perhaps even dramatically so. The stock market didn't fully capture this nuance on Friday, though the bond market definitely did, as interest rates went back up and bond prices went down in dramatic fashion. These rate increases even came in the face of political instability in Egypt. Such turmoil would usually cause a rush to the safety of United States government bonds, increasing bond prices and decreasing interest rates--but not this time. The 10-year Treasury bond jumped to 3.66%, well off its 2010 low of 2.4% and year-end level of 3.3%.

The Official Line: We Gained a Dismal 36,000 Jobs

The official employment report from the U.S. Bureau of Labor Statistics, the establishment survey is usually the gold standard of all economic reports. The report showed meager job growth of just 36,000 jobs in January. The trend for the past four months is as follows:

Why Do I Think the Official January Employment Report Is Flawed?

Weather appears to have been one major cause of the slow job growth. Industries subject to weather issues looked horrible (construction down 32,000, transportation including couriers down 38,000, temporary help down 11,000). Meanwhile bellwethers like manufacturing (up 49,000) and retail (up 28,000) looked incredibly strong. In fact, manufacturing job growth was higher than any other period dating back to 1998. The number of people unable to find work because of weather was nearly triple the usual January average, though that data is from a different survey.

Benchmark Revision Makes Recovery Look a Little Slower

Although it doesn't appear to be a huge factor, the incorporation of actual employment tax records and unemployment databases into a new benchmark revision also affected the data. This in turn triggered new seasonal adjustment factors as well. The net of these changes is that the employment recovery began later and slower but has picked up more steam in recent months than previously thought.

New Business Creation Metric Stings Results
The new business creation and business failure adjustment was gigantic this month: a huge 339,000 jobs subtracted from the job report, one of the larger adjustment factors of the year. Unfortunately, small businesses don't usually mail a postcard to the BLS saying, "My business just failed." Instead, the failed businesses simply don't respond to written or telephonic communications, forcing the government to decide whether the nonresponders are just slow or their businesses went belly-up. Likewise, as new businesses are formed, it takes the government a while to actually find a name, address, and phone number, again forcing the government into estimating. To make things even more fun, the government changed the timing on making some of these adjustments. Usually January is a big month for new businesses, hence the big adjustment factor. However, I suspect that with tight lending conditions, new startups are a little slimmer this January than the seasonal adjustment is implying.

Other Reports Show Stronger Job Growth

However, one doesn't have to look too far to find other reports that show a more robust employment situation. In fact, the news out of the BLS' own household survey showed shockingly good results. These surveys ask citizens (rather than establishments or businesses) whether they are working or not. The surveys often diverge over a short period of time but tend to line up over longer periods of time.

It should be noted that economists tend to shy away from the household report because the unemployed are known to stretch the truth about their employment status out of either fear or embarrassment. That said, the household report showed job growth of almost 600,000 people (as adjusted for a population benchmark change) compared with the miserly 36,000 reported by businesses. In December the household survey showed job growth of 297,000 compared with the establishment survey growth just 121,000. The household data tend to be a lot more volatile than the establishment survey and cover a bigger base. Therefore, I wouldn't stretch the data to say we had huge job growth--just better than the establishment survey suggested.

Likewise, data from ADP, the people who actually print a lot of payroll checks, indicate that the job market is doing better than the employment survey suggests. ADP data suggest payrolls grew by 247,000 in December and 187,000 in January. While the ADP data had underestimated payroll growth by small amounts each month earlier in the year, we have now seen two months of huge overestimates compared with the government data.

The anemic job growth figures don't square well with the initial unemployment claims data that have recently shown some nice improvement, too. Claims have fallen to 413,000 from 458,000 over a four-month period. Fewer firings should mean more employment growth, unless new hires are moving in the wrong direction as well.

Adjusting for Everything, I Believe January Jobs Growth Was Close to 200,000

Putting all three reports together, plus the initial claims data, I think that without the weather and other special factors, employment growth would have been closer to 200,000 jobs. That's still not robust enough to make a really serious dent in the unemployment rate, but it's enough to calm some of my nerves concerning consumer wage income. Making my adjustments, it looks like employment was up, as were real hourly wages. Hours worked showed a small decline, primarily due to weather-related delays in the construction industry.

Inflation-Adjusted Wages Pop

The hourly wage data from January are also not totally consistent with a labor market weakening. I assume that employees would not be able to extract larger-than-expected wage growth if labor markets were in the complete shambles that the establishment survey suggests. Wages jumped over 0.5% in January, and this annualizes to wage growth of over 6%.

While these figures are seasonally adjusted, I suspect some of the calendar year unfreezing of wages may have helped the data along a bit.

Poor hourly wage growth, at least on a month-to-month basis, was the main reason we raised a yellow caution flag several weeks ago. However, the numbers are a bit more benign when we include this month's data and average those values compared with a year ago. Real hourly wages are the best early indicator of consumer wage growth and spending.

Hours Worked Hit by Construction Industry Weather Issues

Hours worked, the third factor to understanding consumer incomes (employment and real wages being the others) was inconclusive this month as a massive weather-related decline in the construction industry caused hours worked to show its first decline since February 2010. Construction hours fell 0.3 hours, or about 1%. Hours worked jumped from a bottom of 33.0 hours during late 2009 to 33.5 at midyear as employers opted to add hours instead of adding employees. However as job growth kicked in, the dramatic improvement in this category came to an end.

News from the Manufacturing Sector Remains Robust
News on the manufacturing front remained solid throughout the world, especially in the U.S. and U.K., as the all-important purchasing managers' surveys for January were released last week. The strong new orders subcategory from the reports bodes well for production and employment in the months ahead. Our industrials analyst Eric Landry summed up the U.S. report as follows:

The ISM manufacturing index broke out to a new cycle high. January manufacturing activity accelerated markedly across the U.S., with the widely followed PMI index registering an impressive 60.8, or 2.3 points above December's 58.5. The latest reading is higher than the previous current cycle high-water mark of 60.4 set in March 2010 and the highest since 61.4 in May 2004. Even better, the internals of the report looked very encouraging. New orders sprung 5.8 points to 67.8 and sit at a level not seen since early 2004, while backlogs leapt 11 points to 58. Exports gained 7.5 points, while imports increased 4.5 points. There was a troublesome increase in the prices component, however, with that category increasing 9 points to 81. This metric also reached a current cycle high, surpassing 78 set in April 2010. As outlined in many of the international manufacturing surveys, material cost pressures remain top of mind right now and are reason for concern for investors.

Auto Industry Powers Ahead

News from the auto front was also surprisingly robust, led by strong consumer demand with less dependence on fleet and rental car activity. Our auto analyst David Whiston explains why he continues to be bullish on the auto sector in the months ahead:

Automakers reported better-than-expected January U.S. light-vehicle sales on Tuesday. Sales increased 17% year over year, making it the best January since 2008. The seasonally adjusted annualized selling rate (SAAR) came in at 12.6 million, the same as in December. Still, the fact that the SAAR continues to rise from last summer's mid-11 million range indicates to us that the auto industry's recovery is on track. We continue to expect U.S. light-vehicle sales to be 13 million this year as we have long argued that 2009's 10.4 million was far below normative demand and the recovery will take several years. The SAAR is still not back to replacement demand, which we estimate to be about 13.3 million vehicles. Furthermore, Automotive News reported last week that the average December wholesale used vehicle price was $9,810, down 0.1% from November and down 1.4% from December 2009. This decline is the first year-over-year decline in 20 months. We think these data suggest that used vehicle prices are peaking and that more consumers will opt to buy a new vehicle instead of an expensive used one. Thus, we expect the SAAR to increase as 2011 unfolds.

Trade Data Due This Week

Things quiet down this week with only the balance of trade report due. I suspect the trade number will move back up again in December after some unsustainable low numbers in prior months. Trade, including strong exports and a surprising drop in imports, powered the strong GDP report previously. The consensus is expecting the trade deficit to move to $42 billion from $38 billion, which seems reasonable. I expect a bigger spike in 2011 as businesses begin to stockpile goods in anticipation of price increases and as U.S. consumers continue to spend.

This is an edited version. The article originated from Robert Johnson’s column at Morningstar.com.