May 06, 2013
Heritage Assesses the Ever-Expanding Ever-Centralizing Federal Government Sector
State and local governments avoided the massive job losses of 2008 and 2009 that affected the private sector—these governments even grew slightly during the recession. But they have been gradually downsizing ever since. The federal government, by contrast, has expanded rapidly since the recession began. Federal employment, excluding the U.S. Postal Service, peaked in 2011 at 13 percent above 2008 levels. At the same time, the private sector was still mired in the slow recovery, 5.5 percent below 2008 levels. Since 2011, federal expansion has stopped, and a fifth of the recession-era expansion has been reversed.
However, most of the federal employment expansion that took place from 2008 to 2011 remains in place. Despite protestations that the additional employment associated with the stimulus would be temporary, federal employment remains as high as it was three years ago and 10 percent higher than it was before the recession. By contrast, private, state, and local employment are 2 percent to 3 percent below pre-recession levels.
The authors use the below figure to illustrate these points.
Here I provide two other ways of examining the data. Figure 1 below depicts employment levels normalized to 2007M12 levels.
Figure 1: Private employment (blue), Federal ex.-postal service (red), and state and local employment (orange), in 000's relative to 2007M12. Federal series excludes temporary Census workers. Source: April 2013 employment release.
Now, it is interesting to observe that should one not exclude postal workers (after all, many other Federal workers are distributed throughout the nation, just like postal workers), then the picture changes somewhat, as shown in Figure 2.
Figure 2: Private employment (blue), Federal (red), and state and local employment (orange), in 000's relative to 2007M12. Federal series excludes temporary Census workers. Source: April 2013 employment release.
In other words, Federal employment is essentially back to levels of 2007M12. I think these two graphs cast in a slightly different light the authors' main contention, viz.:
...The expansion of the federal government has ultimately come to some degree at the expense of states and localities. As the federal government seeks to command more of the economy—most notably the health care sector—this troubling trend is likely to continue to move employment and power to the least transparent and accountable level of government
I guess it is an expansion of the center if the non-center shrinks. (Music side note: I almost expected the Darth Vader theme!)
Finally, I find it interesting that Sherk and Furth do not mention how little government employment grew relative to previous episodes.
Figure 3: Log government employment (ex.-census workers) relative to 2007M12 peak (blue), relative to 2001M03 peak (red), and relative to 1990M08 peak (green). Source: BLS, and author’s calculations.
As I recall, George W. Bush was President during the 2001M01-2008M12, when government employment surged. I do not recall a similar concern about centralization of power on the part of Heritage Foundation economists at the time.
In absolute terms, the change in government employment is 1.6 and 1.5 million lower in the current episode relative to the 2001M03 and 1990M08 episodes, respectively. Interestingly, this figure is very close to the 1.4 million that Chinn, Ferrara and Mignon (2013) identify as the structural change in total nonfarm payroll employment not attributable to misprediction in private employment; i.e., a good chunk of the low employment growth is directly attributable in a mechanical sense to the unique negative trend in government employment. (If you thought the previous two episodes were anomalies, see WSJ RTE.)
In my view, the NY Times assessment by Nelson Schwartz better summarizes matters:
“The drag from the government sector is quite substantial,” said Gregory Daco, senior principal economist at IHS Global Insight. “Given the fiscal headwinds, the private sector is doing O.K.”
“If it weren’t for the government, the economy would be stronger,” said Mr. Daco, citing the spending cuts hitting now, as well as the higher Social Security deductions for all workers and increased income taxes for top earners that began in January.
See this post for a discussion of Messrs. Sherk and Furth's previous interpretation of employment statistics.
Posted by Menzie Chinn at May 6, 2013 10:00 AMdigg this | reddit
I looked at the Heritage piece linked. Can you explain their employment index graph?
Posted by: jonathan at May 6, 2013 10:23 AM
The geographic dispersion reason the authors cite for excluding postal workers is silly.
But they also mention a legitimate reason, the fact that USPS revenues come from market transactions. In this way the USPS is much more like a private sector employer than a government employer. You can be sure if the USPS were funded out of general tax revenues rather than market demand, they would have had far fewer (if any) layoffs.
Posted by: W.C. Varones at May 6, 2013 10:41 AM
It is sad that economists today are driven by political agendas rather than truth. Unemployment has been brought down because the permanent jobs have been turned into part-time jobs. The government has discovered that two people working 20 hours gives them better propaganda than one person working 40 hours, but are we really better off.
But then on top of that there has been a huge exodus out of the work force. Many blame the baby boomers retiring but truthfully the massive growth in SS disability roles has a significant impact and where those on unemployment are still in the workforce those on SS disability are not counted in the workforce.
The latest jobs report is just more smoke and mirrors.
Posted by: Ricardo at May 6, 2013 12:19 PM
Comment #1) Menzie's expert from the NYTimes article is misleading. He neglects to included the subsequent sentence: "On the other hand, he said, “If the Fed hadn’t loosened monetary policy, we’d be seeing weaker growth. Both sides are generating opposing forces.” Recognizing this monetary offset is absolutely critical to any policy discussion. Menzie seems to give it short shrift either because he doesn't fully understand it or because it doesn't support his constant appeals for more fiscal stimulus.
Comment #2) If one accepts Menzie's characterization that a "good chunk of the low employment growth is directly attributable in a mechanical sense to the unique negative trend in government employment", then doesn't that imply that those workers were completely unproductive? And yet, additional (non-productive) fiscal stimulus is needed? Oh the irony!
Posted by: Jeff at May 6, 2013 12:34 PM
I appreciate you trying to hold President Bush up as a standard for President Obama's growth in federal government employment, but I have to tell you it rings hollow with those of us who believe in a liberal constitutional, democratic, republican form of government. We opposed Progressive government under Bush just as much as we oppose Progressive government under Obama, so we don't find an appeal to the Bush policies very convincing.
I admit that Heritage went through a struggle with its objectivity concerning the Bush administration and in large part that is why Jim DeMint was brought on board, but in truth there were may liberal publications opposed to Bush's Progressive economic policies. Here are just a few.
Posted by: Ricardo at May 6, 2013 01:08 PM
@Ricardo: Disability has not "exploded" outside of the simple demographics of an aging baby-boom population. http://www.washingtonpost.com/blogs/wonkblog/wp/2013/04/01/the-simple-boring-reason-why-disability-insurance-has-exploded/
Posted by: RWForce at May 6, 2013 01:12 PM
jonathan: What they did is to generate an index based at 100 for 2008M01 (so, X[t]/X[2008M01])×100). So small changes to a relatively small number look big. Not "wrong" per se, but it's not the way I would do it.
W.C.Varones: USPS borrows from Treasury to cover shortfalls. See this CRS report.
Ricardo (12:19): You write "The latest jobs report is just more smoke and mirrors." I'm surpised you didn't break the story that the moon landings were faked, too.
Ricardo (1:08): Yes, I know there were some voices in opposition; my point was merely to point out the inconsistency at Heritage, after I did (an admittedly) cursory search through there online documents.
Posted by: Menzie Chinn at May 6, 2013 01:48 PM
Here's a more detailed analysis of Executive Dept federal govt employment, by department/agency.
Overall federal employment has been flat to slightly downward trending over the last 45 years. Most govt agencies have seen falling employment. The three exceptions are Defense, Homeland Security and Veterans Affairs. Given all the wars we've had over the last dozen years I don't think this ought to come as a big surprise. And recall that Homeland Security grew because it picked up all the former minimum wage private sector flunkees who were asleep at the switch. Say what you will about those TSA folks...and I grumble as much as the next guy, but they're a hell of a lot better than the clowns they replaced.
Most of what government produces falls in the category of "public goods." And the market demand curve for public goods is derived by vertically summing all private demand curves for public goods. So you would expect the demand for public goods to increase with population, and particularly with population density. So why would anyone expect government employment in 2013 to be less than it was when LBJ and Nixon were President? Especially if government workers are as unproductive as many seem to believe.
W.C. Varones Unlike UPS and FedEx, the USPS doesn't get to choose which markets it will serve and which ones it won't. Unlike UPS and FedEx, the USPS doesn't get to choose how much to charge. Unlike UPS and FedEx, the USPS actually has to fund its retirement programs. Unlike some private sector delivery firms, the USPS does not require (or even allow) its drivers to purchase the capital equipment from the company as a condition of employment. And unlike UPS and FedEx, the USPS doesn't get favored treatment from the federal government when it comes to the lucrative overnight and express deliveries.
Posted by: 2slugbaits at May 6, 2013 03:10 PM
slugs: Sometimes I think you just cut-and-paste things from other sources without understanding them. Employment is an input for the supply, not demand of goods. So the public good argument is an argument for why public employment will not scale with population.
Posted by: Jeff at May 6, 2013 04:11 PM
Jeff Perhaps I wasn't clear enough. The demand for public goods increases with population. If public employment is virtually flat, what does this say about govt worker productivity? Increasing output with no meaningful change in labor inputs.
Posted by: 2slugbaits at May 6, 2013 04:28 PM
Jeff: Well, I thought everybody reading this blog knew about QE1-3, but I guess I overestimated the knowledge of some readers.
You can't make the conclusion you do because GDP includes defense, and government employment includes only civilian. Some day, you should learn some national income accounting.
Posted by: Menzie Chinn at May 6, 2013 05:08 PM
Did you read the article you referenced. Her thesis is based on an increase of 5%.
Congress has hiked the retirement age for Social Security over the years — in fact, about 5 percent of current disability recipients would simply be on ordinary Social Security had Congress not changed the rules.
That said, demographics aren’t the whole story. The program is still growing even after controlling for age and gender dynamics.
Check out this more indepth analysis.
Posted by: Ricardo at May 7, 2013 05:49 AM
I know you deal repeatedly with idiot commenters but I'd hope you give credit to being able to read the actual labels on a chart.
My question wasn't the dumbest you could answer - i.e., how do you scale a chart - but what is the number made of?
I assumed you'd know the composition of federal employment, that you didn't just run some series. Composition is the question posed by the content of the graph: it excludes what? postal employees. So what does it include? Answer is homeland security, DoD, veteran's affairs. These are all up since 2008. Everything else is flat. Well, there's a small increase in HHS. Outside of the first 3, the ratio of federal employees to population has fallen. It may have fallen with those 3 included but I didn't check that.
My experience with Heritage is they manipulate the content. You focused on the graph. It's accurate. Misleading but accurate. But they make a point about centralization and federal growth and that's more than misleading because the areas where that growth has occurred is where Heritage supports. They use that near lie to imply wrongly that federal employment in HHS, Education, etc is burgeoning. It isn't. The only thing growing is defense and security.
Maybe sometimes you're a bit too numbers oriented.
Posted by: jonathan at May 7, 2013 07:15 AM
Menzie, would comparisons of previous recessions and government employment be improved if you used both federal employees per capita ex military ex post office, state and local over time as well as your Logs = 0 at trough. I agree with your assessment.
See my http://www.clintburdett.com/articles/cbscminiblog.htm
Posted by: Clint Burdett at May 7, 2013 07:24 AM
jonathan: Apologies. I thought it was a serious question, since it seemed like such a disingenuous thing that Sherk/Furth did, that you could be excused for asking if they really did what they did. Now I see what you're getting at; yes, I agree, it's also appalling for the reason you mentioned. (Also, yes, sometimes I'm too numbers-focused.)
Posted by: Menzie Chinn at May 7, 2013 07:34 AM
One may find the M2 velocity an old saw, but coupled with actual employment numbers the actual money velocity is an oddity or the employment numbers do not match.Yet there are papers revolving around money velocity and employment.
Needless to look in the ECB data wharehouse for the same, an old saw.
Posted by: ppcm at May 7, 2013 08:53 AM
2slugbaits, thanks for that link - that's good data. Like you pointed, most of the growth in Federal government jobs has been Dept. of Defense, Homeland Security, and Veterans Affairs.
In fact, of the 186,000 jobs added to the fed payroll since 2008 (using 2009 might be a better comparison since 2008 was still the previous administration), 100,000 are DoD (56%), 22,000 are Homeland Security (12%), and 40,000 are the VA (21%). Those are all agencies that, if cut, the immediate reaction among the right is "he's gutting the military!", "the terrorists are going to win!", or "he's not supporting our troops!"
Posted by: MV at May 7, 2013 09:34 AM
Dec-07 Apr-13 Amount %
Fed 2,756 2,769 13 0.5
State 5,139 5,053 -86 -1.7
Local 14,481 14,022 -459 -3.2
Total 22,376 21,844 -532 -2.4
Fed(F) 2,756 2,769 13 0.5
Post)P) 782 590 -192 -24.6
F-P 1,974 2,180 205 10.4
As Jonathan said the Heritage data appears correct, but perhaps gives an incorrect slant.
Posted by: AS at May 7, 2013 10:19 AM
Menzie: It's a tad more than just being aware of it. It's understanding it and understanding it's implications, which you have consistently demonstrated a lack of.
Your snark about accounting identities is misguided. You suggested that the drop in federal (civilian) employment captured a large chunk of your estimated of structural unemployment. I then logically extended that claim to make the observation that one must conclude the federal (civialian) unemployed must be non-productive. I suggest you spend a little more trying to understand what you read.
Posted by: Jeff at May 7, 2013 11:46 AM
slugs: But you assumed the goods were public goods, which means that they are non-rivalrous, which means no increase in output.
Posted by: Jeff at May 7, 2013 11:49 AM
I must wonder with incredulity why on earth anybody even gives the Heritage Foundation's analysis of anything the time of day, much less spends the time to debate their junk.
The only people that believe what-ever the Heritage Foundation writes and take it as the "gospel" and last word are those that believe gov't is always larger and has more control than it should have... no matter what size or magnitude of control it has. At their core they are fundamentally anarchists.
Posted by: longtooth at May 7, 2013 01:26 PM
As to government employment the question that must be asked and answered isn't whether it's now higher or lower than it was at some prior point in time, but rather what should the number be at any point in time?
When it comes to government employment numbers the far right always says it too high... too much gov't inefficiency, lazyness, waste, etc. When I ask what the gov't outlay on government employment should be the only response I get is 'less than it is now'. Not very objectified to say the least. The always exception cited is that defense should be larger!
This is more than just a little bit obvious that the objective has nothing to do with how many employees the gov't has, but rather it's simply is to reduce central gov'ts power.
To what ends? To the point that each state is a sovereign unto itself of course, with an interstate agreement to use the federal gov't only as their (the states') agent for defense only. Something on the order of the Articles of Confederation seems to be the fundamental objective.
Most everything else is just propoganda & rhetoric intentionally designed to mislead and obfuscate the real objectives.
BTW, once the state sovereignty objective was attained the next objective would be county sovereignty, and it degnerates from there right on down to anarchy.
Posted by: longtooth at May 7, 2013 01:55 PM
Jeff Your understanding of public goods is incomplete. What makes a good a public good is the non-rivalrous nature of the consumption of the good, not the cost curve of the good. There is nothing about public goods that says the marginal cost of production must be zero. Defense is a public good, but the marginal cost of producing more defense is certainly not zero. Broadcast television is a public good because consumption is non-rivalrous. That does not mean the marginal cost of shows produced for broadcast are all zero. Better shows cost more to produce. Remember, the market demand curve for a public good is derived by vertically summing private demand curves. That means the intercept moves up, which shifts the demand curve. As population increases the intercepts for potential public goods will eventually shift high enough that production of new public goods makes sense. So not only will the quantity of public goods demanded tend to increase, but so will the scope of public goods. In other words, the scope of government will tend to increase as society gets wealthier and populations increase. Of course, as society gets wealthier people also want to demand more private goods. But they will also want more public goods for the same reason; i.e., an income effect.
Posted by: 2slugbaits at May 7, 2013 03:44 PM
Jeff (11:46AM): Sigh. OK, suppose the true relationship between civilian nonfarm payroll employment and nondefense output is given by n = αz and total GDP is given by y, so z = y-m (where m is defense output). Then the true relationship is n = α (y-m) = α y - α m. Then if m trends differently in a particular sample -- say the out of sample period, it will look like nfp is unnaturally low versus the long run relationship. In point of fact, defense spending in the 2008-12Q3 period was 0.8 ppts higher than the preceding 1990-2007 period. It's not conclusive proof, but the fact that one could get the result we found, without resorting to zero productivity government workers, was what I was trying to get at.
Posted by: Menzie Chinn at May 7, 2013 07:29 PM
Ricardo: You are certainly correct that there was a shift from full-time to part-time employment (involuntary short hours, or rather working part-time for economic reasons). However, the level of such underemployment is tracked by BLS and reported in their extended unemployment series (specifically U-6). The level is still high, but it too has fallen. See a graph HERE.
Posted by: mike smitka at May 7, 2013 07:37 PM
More on Disability Insurance.
Posted by: Ricardo at May 8, 2013 06:11 AM
In April U6 increased.
Posted by: Ricardo at May 8, 2013 06:13 AM
Ricardo: Context is sometimes useful. U6 increased by 0.1 ppts. It declined from 14.4 to 13.8 from Dec 2012 to March 2013. See here.
Posted by: Menzie Chinn at May 8, 2013 08:13 AM
Slugs: No my understanding of public goods is not incomplete, you are conflating issues. You are lumping the issue of non-rivalry with a basic income effect. Take your defense or TV examples. For a fixed amount of defense or TV, an increase in the population does not detract from everyone else's consumption of defense or TV. More people will demand the goods and more people will be consuming the good but no additional supply is required. Note I am not assuming an increase in incomes, only population. That is the essence of non-rivalrous. To argue that more TV stations or larger defense will accompany larger populations because countries have tended to get richer and thus demand more goods may be true in reality but that is because of a general income effect and not due to the public good nature of goods.
Posted by: Jeff at May 8, 2013 10:12 AM
Menzie: Well of course there are always going to be other possible reasons for why we observe certain observations. That's a tautological and tremendously insightful point. All regressions deal in averages, so to simply appeal to possibility that a certain sub-sample may be driving the result is rather pedestrian.
Posted by: Jeff at May 8, 2013 10:29 AM
Jeff: OK. Well, I've never minded being characterized as pedestrian. Better to be pedestrian and correct than provocative and wrong. Which makes me wonder -- why do you keep on reading my posts? Don't you have something better to do than comment on pedestrian posts?
Posted by: Menzie Chinn at May 8, 2013 10:37 AM
Jeff: "But you assumed the goods were public goods, which means that they are non-rivalrous, which means no increase in output."
I don't understand what you are trying to say. Are you suggesting that an increase in population should not result in an increase in government (cost? output?). I don't understand that. Why wouldn't it be reasonable to assume that e.g. a population increase of 10% will result an increase in the need for e.g. number of teachers needed or number of DMV employees needed to issue drivers licenses?
There might be some instances where population increases doesn't necessarily increase the amount of government "good", defense comes to mind.
However in all of these instances the "consumer" of the good/service is not individuals but rather The Nation, of which there is only one. In fact I can only think of Defense in this category, and even there it's iffy, since defending 200 population centers could reasonably be expected to cost more than defending 50 cities. Also, we shouldn't forget that our defense forces has been extensively used to secure economic resources for the nation (e.g. shipping lanes, oil, trade), and it's reasonable to expect that the number of entanglements should grow with the economy, which grows with population.
Posted by: Anonymous at May 8, 2013 11:58 AM
Menzie: You give yourself too much credit to say you are pedestrian and right. I usually comment when you are wrong.
Posted by: Jeff at May 8, 2013 01:59 PM
Jeff Let me try one last time. Adding one person to the population moves the intercept up irrespective of any additional income effect, which is a shifting of the curve. For example, if there are initially 10 people in society, then the market intercept is the vertical summing of each of the 10 individual private demand curves. If an 11th person comes along, then the intercept gets shifted up again by an amount equal to that person's intercept for that person's private demand curve.
It might help if you drew some curves to observe what happens to the demand and supply equilibrium as you shift up the intercept for the demand curve. Start off with a supply curve that is much higher than any individual person's demand curve. If you add enough private demand curves, then eventually the market demand curve will intercept the supply curve. That's an example of population growth expanding the menu of public goods that should be provided by government. Add more demand curves and the amount demanded will increase as well. But you have to keep in mind that public good outputs typically have a different meaning than private good outputs. It's not meaningful to just talk about "defense" being a public good. It's really multiple degrees of "defense." Do you want to have enough defense to fight one small war? Two small wars? Crush every imaginable conflict scenario? That's what we mean by the amount of defense being provided as a public good. And that will tend to increase with population. Now for any given "degree of defense" the marginal consumption of the good is zero; but the marginal cost to produce each "degree of defense" is positive.
Posted by: 2slugbaits at May 8, 2013 03:28 PM
Jeff: Well, I use your comments as an indicator that I've got something right -- so we're even. But you didn't answer my question -- don't you have something better to do than to read my "pedestrian" posts? Mebbe publish a paper somewhere? I'd look forward to reading one of yours.
Posted by: Menzie Chinn at May 8, 2013 03:32 PM
Since many government functions are contracted out, how does this fit into federal employment? For instance Booz Allen and Hamilton, SAIC, Battelle Memorial Institute, etc. have divisions and department that are exclusively federal contractors. Such as running national DOE labs or professional consulting. Not to mention the many contractors providing services to the DOD, where revenues are exclusively based on federal dollar flows. Anyway great blog. I just starting reading.
Posted by: Michael Berndtson at May 9, 2013 03:59 AM
Slugs: You're still re-framing the problem and moving away from your original point. You are simply talking about a general income effects of goods that happen to be public goods. There is nothing in the what you are currently describing that is dependent on the goods being public.
Posted by: Jeff at May 9, 2013 12:48 PM
Menzie: Yes, your one intro-to-econometrics comment that regression coefficients are average effects was correct. If an economics professor at a top 25 school takes pride in being right on that point, then who am I to judge. And yes, there are certainly more productive things to do that read blog post, but I suppose I come for the entertainment value.
Posted by: Jeff at May 9, 2013 12:56 PM
Jeff: Sorry to disappoint you, then. Please continue to read. You provide equal if not more entertainment for me. I look forward to your next defense of Heritage analysis.
Posted by: Menzie Chinn at May 10, 2013 08:36 AM
What I would view as an important point is that absolute numbers of government employment regardless of composition are not the best indicators. What should be done is to view government employment as a factor of both potential GDP and Population. Also some mission analysis is important. Government employment grew with the defense expansion and also grew with the social security mission. The real question should always be what are they doing. Governments can vary their civil service employment by hiring support contractors. Whether this is a good idea or not is not the immediate question but that it bears on understanding the numbers. For example, what was previously done for base support in the Defense Dept is now largely done by contract workers. This now changes the composition and numbers of civil service workforce. Interestingly, a similar item has occured in the military where logistics contracting e.g food services has been substituted for military personnel-- so much for KP.
Posted by: Eclectic Obsver at May 10, 2013 10:56 AM
Menzie: Exactly where did I defend Heritage analysis? Or is this Chinn's version of Godwin's law? Eventually he'll call you a Republican.
Posted by: Jeff at May 10, 2013 06:58 PM
Jeff: Apologies, it must be another Jeff, who argued that the idea that one dollar taken away from government spending is immediately re-employed in the private sector. You both had such a similar tone of argument, I mistook you to be the same person.
Posted by: Menzie Chinn at May 11, 2013 08:27 AM
After reading this thread, Ive reached the conclusion that whatever the merits of his arguments, this Jeff is an arse-h*!e. Im not particularly impressed by the force of his arguments, but that wouldbe for someone more knowledgeable than I am to say.
Posted by: outside ovserver at May 11, 2013 08:08 PM