Running Government Interference

by Brian Pretti CFA, Contrary Investor. October 30, 2009

A few very quick comments about recent trends in payroll employment that just don’t seem to be getting enough attention as of late, especially given the clear change we are seeing over the last six months or so. You may remember that probably close to a year ago I penned a discussion focusing on the rhythm of government employment looking back many decades. The important point of that discussion was that the rhythm of government employment lags that of the private sector in each economic cycle. In other words, private sector employment turns down long before government employment begins to weaken in each cycle. Wildly enough, government employment trends tend to weaken late in recession cycles and continue soft well after official recessions conclude. Not too hard to understand in that it’s during the midst of economic contractions that income tax, property tax, retail sales tax, etc. revenues derived by all government entities decline meaningfully, as has exactly happened again in the current cycle, but this time with a vengeance.

It’s when this occurs that government entities at all levels begin to tighten their belts and employment is negatively affected. Well guess what? Although it seems there has been virtually zero commentary about this, since May of this year just as the “green shoots” crowd began crowing about payroll employment getting less bad, almost right on cue government employment trends began to break down. In nominal body count terms, aggregate government payrolls (Federal, state, and local combined) peaked in April and have declined by over 200,000 through September. In the five-month period ended September, 213,000 government jobs have been lost, the bulk of which come at the State and Local government levels. Here’s a fun stat for you to amaze and dazzle your friends at the next cocktail party. Since 1969 there have only been two other five months periods in which government employment has shed as many jobs as has occurred in the past five months – September of 1980 and October of 2000. To suggest this is not a common occurrence is an understatement. The following chart documents the nominal body count and year over year rate of change history of government employment in aggregate. As the chart shows us, the current year over year rate of contraction has no equal until we reach back to 1982.

Funny, I have not personally seen this emerging trend in government employment weakness talked about anywhere. This is something to keep in mind as we move ahead, because it is now when the government sector in aggregate could become the next drag on overall payroll employment. You know that my wonderful home state of California squeaked by passing a so-called budget some month’s back. The budget was a band-aid. It simply kicked the can of hard decisions down the road into 2010. Bigger budget cuts are to come by necessity. And they will virtually certainly involve explicit payroll reductions next year. There is little other alternative. What California has done so far is to declare “holidays” a number of times per month for many State employees. Effectively they have already cut payroll in concept, but it has not shown up in official payroll numbers as these folks are still employed; they just work less hours each month and collect less pay. California is not alone in this poster child example of what is occurring in many states across the country. So, is it fair to say that for many States and municipalities, payroll headcount is currently overstated given that hours worked and nominal dollar payrolls have been cut? Seems more than fair, no? But given the rhythm of historical experience in government employment set against economic cycles, it is exactly now that we need to expect and anticipate the government sector becoming a drag on overall payroll headcount over the remainder of this year and into next. Be prepared as history tells us this is exactly what is coming.

Although this may not be common knowledge, the bulk of government employees are found at the State and local levels of government. In fact, Federal government employment only accounts for 12.6% of total aggregate government employment as of September 2009-month end. That’s pocket change compared to State and local levels of employment. And of course the current cycle irony is that it’s the State and local governments that are hurting big time under the duress of not only infrastructure, maintenance and support, but pension issues, loss of Federal support payments, etc. The following combo charts puts into perspective for you what local and state employment trends have looked like over the last four decades. Current rate of change weakness relative to prior cycles is indeed meaningful and important to keep in mind ahead. Something no one else appears to be concerned with at the moment.

So why bring up government employment now and why is it important? First, throughout the current recession until now, the government sector has been a net job creator, so this is change at the margin in an already weak macro labor market environment that appears very worthy of note. As of September month end, total government employment accounted for 17+% of total US employment. C’mon, this is not the end of the world at that kind of a level relative to the total, right? Well, here’s just a bit of perspective for you in just how important government employment has become over the last few decades relative to other sectors of the economy such as manufacturing. Have a look.

As of now, unfortunately for the US economy as a whole, government employees outnumber US manufacturing sector employees literally just shy of two to one. Now is the rhythm of government payrolls important enough for you? Do not be surprised if government payroll trends become a further drag on the headline numbers ahead. As a final comment, the following table delineates just how far after each recession end the annual rate of change in government employment bottomed. Just a frame of reference.

Recession Ends	Annual Rate Of Change Trough In Government Employment
11/70	4/71
3/75	10/76
7/80	2/82
11/82	2/83
3/91	5/91
11/01	1/04

Although I did not calculate the average time lag in this table, on average, the rate of change trough in government employment occurs 15 months after each recession conclusion. In our current circumstances, that would imply a rate of change low at year end 2010. We'll just have to see what happens ahead, but given the dynamics of the current cycle and the fiscal circumstances of State and local government entities, forward character of government employment may resemble a long and slow bleed. Have investors factored this into decision-making? Personally, I’m not so sure.

Brian Pretti

© 2009 Brian Pretti

Contact Information

Brian Pretti CFA | Editor and Publisher, Contrary Investor
P. O. Box 4402 | Walnut Creek, CA 94596 | Observations | E-mail