I thought I’d add a few observations on the latest employment report (other reports here: [NYT], [WSJ RTE/Izzo] [CR], [Economist’s View]). First, by an alternate measure, employment is improving more rapidly than the standard nonfarm employment (NFP) measure. Second, the alternate measure increased faster than nonfarm payroll employment over the period of temporary Census hiring. Third, aggregate hours worked in the private sector continues to rise faster than private sector employment. Fourth, the NFP growth consistent with zero GDP growth is lower in the last decade, versus previous decades, even while the elasticity of NFP growth with respect to GDP growth has risen.
The Standard and Nonstandard Time Series
Figure 1 illustrates nonfarm payroll employment, and the same series excluding temporary Census workers.
Figure 1: Nonfarm payroll employment (blue line), NFP excluding temporary Census workers (red line), and civilian employment series adjusted to conform to NFP concept (green line), all seasonally adjusted. Gray shaded area indicates recession, assuming trough at 2009M06. Source: BLS via FRED, NBER and authors calculations.
As described in earlier posts [1] [2], the adjusted civilian employment series is constructed to conform to the nonfarm payroll series concept, but based on data from the household survey. One characteristic of this research series is that it has higher variability, measured as the standard deviation of the log first difference, than the nonfarm payroll series, so one might want to downweight its importance relative to the NFP measure.
Interestingly, the adjusted civilian employment series shows a bigger jump than the NFP series. The elasticity of adjusted civilian growth with respect to nonfarm payroll employment growth over the 2009M12-10M08 period is 1.3. Over the entire 2005M01-10M08 period, the elasticity is 0.88. It could be that the extra spending associated with the extra Census workers induced greater employment in firms not well covered by the establishment survey; or it could just be an artifact of measurement error.
Employment versus Hours Worked
Figure 2 illustrates the nonfarm payroll series, the (log) private payroll series, and aggregate hours index for the private sector, all normalized to zero at 2007M12.
Figure 2: Log nonfarm payroll employment, seasonally adjusted (blue line), private sector employment (red line), private sector aggregate weekly hours index (green line), all seasonally adjusted, normalized to 0 at 2007M12. Gray shaded area indicates recession, assuming trough at 2009M06. Source: BLS via FRED, NBER and authors calculations.
Private aggregate hours are rising more rapidly than (private sector) employment (this characterization remains intact from last month’s post on the employment situation.
Pedagogical Moment / Employment-Growth Stylized Facts
In comments to a recent post, Econbrowser reader CoRev wonders if an increase in GDP necessarily implies an increase in employment (this is in the context of whether an increase in GDP due to ARRA necessarily induces an increase in employment). Most theories I know suggest this is true (although, I suppose that if the elasticity of substitution between labor and capital were sufficiently high, one could get inverse movements between the two series). But this is an empirical question in the end. So here is a slight digression on the subject of how linked are GDP and nonfarm payroll employment. Figure 3 below shows a scatterplot of the first log difference of NFP against first log difference of real GDP. The blue dots pertain to 1967Q1-85Q4; red dots to 1986Q1-2007Q3; and green dots to 2007Q4 to 2010Q2. Interestingly the (short run) elasticity of employment growth with respect to GDP growth is slightly higher in the most recent period compared to the 1967Q1-85Q4 period.
Figure 3: Log first difference of nonfarm payroll employment against log first difference of real GDP (both seasonally adjusted); blue circles, 1967Q1-85Q4; red circles, 1986q1-07Q3; green squares, 2007Q4-10Q2. All monthly NFP figures converted to monthly by arithmetic averaging. Source: BLS via FRED II, BEA 2010Q2 2nd release, and author’s calculations.
There is a clear positive correlation, with a slope coefficient of 0.4 for the 1967-85 period, to about 0.5 for the 1986-2010 period (statistically significant at the conventional levels).
What changes is the intercept in the regression of first differences on first differences; it’s lower in more recent periods. A similar pattern is observed with respect to year-on-year changes. In other words, at zero GDP growth, employment growth is now lower (essentially 0%, post 1985, -0.2% post 1999) than it was in the 1967Q1-85Q4 period (0.23% per quarter, all figures not annualized).
A more formal (statistically speaking) treatment of the link between employment and GDP is in this post. Hence, for reader CoRev‘s benefit: (1) there is a short run correlation between GDP growth and employment growth; there is a long run relationship (in a statistical sense, called cointegration) between log GDP and log nonfarm payroll employment. If GDP is increased due to say an increase in government spending on goods and services, then in general nonfarm payroll employment will also increase.
Update, 10:15AM Pacific, 9/5/10: Here is a scatter plot of log first difference of hours against GDP.
Figure 4: Log first difference of private sector aggregate weekly hours index against log first difference of real GDP (both seasonally adjusted); blue circles, 1967Q1-85Q4; red circles, 1986q1-07Q3; green squares, 2007Q4-10Q2. All monthly NFP figures converted to monthly by arithmetic averaging. Source: BLS via FRED II, BEA 2010Q2 2nd release, and author’s calculations.
Menzie, I’m not sure I appreciate the special attention. One can never tell what response will come from the peanut gallery.
I agree with your conclusion, with particular attention to the in general caveat. I’m sure both of us can envision circumstances where increased GDP results in improved productivity and no or no statistically discernible increase in NFP.
Anyway, this is not a subject to which I normally comment. I responded just because I was singled out. That was your goal wasn’t it? :))
Menzie “First, by an alternate measure, employment is improving more rapidly than the standard nonfarm employment (NFP) measure.”
Along similar lines, last week I read a new NBER paper that argued the often heard comment that small businesses are the main source of job creation is a stylized fact that doesn’t hold up to scrutiny. It turns out that the sources of robust job creation are young firms, and that among mature firms there is no difference between small and large firms. Evidently youth counts more than size.
http://www.nber.org/papers/w16300
So it’s possible that there’s more hiring going on than we think but those hirings are censored from the establishment survey
“If GDP is increased due to say an increase in government spending on goods and services, then in general nonfarm payroll employment will also increase.”
“In general”, yet not a given. It depends a lot on what kind of spending, capacity utilization, inflation, and business confidence in future business prospects. A lot of govt demand will be supplied with imports. More govt spending will also likely result in higher oil prices, which drains away more of that money to inflation and higher imports. If capacity utilization is low, then hours worked can be increased which eliminates the need to increase net employment. If business confidence in the real economy remains low, businesses will save the money and increase cash levels, rather than expand employment.
Lamont That’s something that businesses can do, but only over the short-run. That’s the signficance of Menzie’s reference to a cointegrating relationship between employment and GDP growth. In the short-run each variable may be nonstationary and follow what looks like a random walk, but over the longer run there is a stationary (and linear) relationship.
And we’re starting to see some evidence that businesses have pretty much exhausted a lot of those short-run strategies to defer new hiring. For example, on Friday the BLS reported that revised 2nd Qtr labor productivity actually fell at a 1.8 percent annualized rate.
http://www.bls.gov/news.release/prod2.nr0.htm
Menzie, when I look at the green dots on your chart it looks like there the positive correlation between changes in emp and gdp breaks down if you restrict your attention to positive changes in gdp(2007-2010). Isn’t that the period corev that sparked corev’s comment? I would expect a stronger positive relationship to re-emerge as we get further into the recovery and uncertainty regarding the sustainability of the recovery and policy is resolved.
[A]t zero GDP growth, employment growth is now lower than it was in the 1967Q1-85Q4 period.
Perhaps a measure of the effect of technology? Well, not solely technology but technology on the then existent economy. In a more manufacturing rich economy, increases in technology increases employment by the effect of the higher personal spending in the national economy.
But today in a services rich economy, an increase in technology just reduces employment by increasing labor efficiency. After all, a goods economy is able to export at a higher level than a services economy.
So a services economy is doomed to higher long term unemployment unless it can find a non-technology driver to goose GDP (like a bubble of some sort?).
Wow, that’s a depressing train of thought.
The debate over stimulus needs a wider context to be useful. Stimulus proponents argue the stimulus “wasn’t enough”, and therefore it is no wonder that recovery is in doubt. But if they knew it wouldn’t be enough, why support it? The answer that it “was better than the alternative” ignores the broader context.
The economy is now attempting to recover in the face of contractionary fiscal policy: the stimulus, as it winds down, will be come a headwind to gdp growth in coming quarters. What is the effect of such a drop in AD on business and household confidence in the midst of a weak recovery? Animal spirits, as always, are at work, and the possibility of an stimulus-induced inventory overbuild cannot be ignored as it may provide the kindling for a double-dip. Further, the failure of Stimulus II has made even the word “stimulus” a political non-starter. What is the long term effect of removing from our arsenal what its proponents believe to be a major weapon?
The easy response to the above argument is the counterfactual: no stimulus would have brought about another depression, so even $800b was better than zero. That ignores the fact that we may land in a similar situation as in early ’09 just two years later, without the possibility of stimulus because it is infeasible politically. Two years is a lot of long-term unemployed, and in the end, from a social perspective, the long-term unemployed are the source of structural economic and social damage to our country.
Lastly, for those still reading, I note the BLS makes an estimate of small business net firm creation through the birth-death model. In all likelihood, the model is now overstating the amount of jobs coming from net creation.
TJ, I actually think there is a strong correlation between GDP and jobs. Given that, simple mathematics says a $800B+ stimulus should create ~7.5M jobs. WE also can go to the Stimulus tracking page and see that only $201B has actually entered the economy. (BTW, much of that was in tax cuts/rebates/incentives.) We can see from Menzie’s chart that jobs have increased ~1.5 -1.75M, and that is in the range (1.8M) that simple math would have predicted.
My question/claim has been why was the stimulus package back loaded? My answer is because of pure election politics.
Why are we asking for a new and bigger stimulus for jobs when the answer is to move forward that back loaded spending? Yes, for me that means any new stimulus spending should come out of the old stimulus package.
My political question for Menzie and the Dems here is: Why did the Democrats, the compassionate party, knowingly extend the pain and suffering for the unemployed?
CoRev: At some frequencies and at some times, one might see a negative relationship, given we live in a stochastic world.
tj: I was focusing more on &partial;y/&partial;nfp , which is the slope of the line through each “cloud” of observations. I find it interesting that this slope coefficient is even larger during the recent period (and was even statistically significant in the 2007Q4-10Q2 period).
GregL: Perhaps it is a different way of phrasing things, but the stimulus package is not exerting a contractionary effect on the economy; if there there was no stimulus spending starting in 2010Q3, the level of GDP would be much lower than that occurring in 2010Q3, while with the ARRA, GDP is likely to be higher in 2010Q3. It’s because the decline in the amount of stimulus, conjoined with lags, is now now deducting from growth.
CoRev asked why the stimulus package was “backloaded.” The answer is in the math. GDP growth reflects rates of change, so as the size of the economy increases you have to have a larger portion of the stimulus reserved for the the latter years just to maintain growth. If it wasn’t backloaded, then the recovery would collapse after the initial impact. Another reason for the backloading is that 40% of the stimulus was in tax cuts. The upside of tax cuts is that they can be started quickly…just a simple adjustment to the withholding tables. The downside of tax cuts is that a lot of the stimulus is a lagged function of income growth. One year after a tax cut you only get one-half the stimulative effect….again, that’s just due to the math.
David Pearson I’m not following your argument. Politically the choice was binary…you either supported passage of ARRA as it was or you opposed it. That’s life. I wasn’t thrilled with the smallball scope of the stimulus package given the magnitude of the problem, but the alternative was Eric Cantornomix, which is to say brain dead economics. The combination of TARP and ARRA at least prevented Great Depression II, which is not a trivial accomplishment. 10 percent unemployment is bad, but without ARRA we almost certainly would have reached 13 percent.
CoRev I’m not sure where you’re getting that $201B figure. Funds (contracts, loans and grants) awarded thru end of March amounted to $202B, so maybe that’s what you had in mind. The Recovery.gov (not to be confused with Recovery.org, which is a right-wing sight intended to confuse the inattentive), total funds paid out to date is $512B.
2slugs, the current Recovery.gov site says that the contracts, loans and grants funds paid out is $145B. Where are you getting the “awarded” number?
I looked at Recovery.gov last week, and that’s where I got the 201B, but I think I probably just sited the fund paid out tax benefits number at that time. My faulty memory at work again.
You did say this: “If it wasn’t backloaded, then the recovery would collapse after the initial impact.” From the Deutsch Bank chart Menzie provided here: https://econbrowser.com/archives/2010/09/what_kind_of_mo.html, we can see the economic impact of the stimulus payments. But, when we compare quarterly GDP numbers we can see the stimulus has provided little economic sustainability.
It is because of this I say the stimulus was misdirected. We did get a stimulative effect early on, but it was too short to maintain economic growth, and that is why I complain about it being back loaded.
Where was most of the incremental growth of private employment since ’98-’00 (when private employment growth peaked, not coincidentally at the end of the previous secular bull market for stocks and the end of the growth boom of the IT-internetworking techno-economic S-curve)? Answer: The FIRE (finance, insurance, and real estate) sector, including the ancillary sectors associated with the peak rate of change of increase in debt/income and debt/GDP.
The 20- to 25-year growth in the FIRE sector (and self-reinforcing effects on IT and other FIRE-related business services) coincided with the peak rate of change effects of Boomer peak earning, spending, and wealth accumulation.
Now the worst 5- and 10-year returns will occur for stocks as the secular bear market enters its second half, coincident with the reverse leverage effects of the debt-deflationary Schumpeterian depression phase of the Long-Wave Trough (average length of 6-7 years historically).
The peak negative demographic cycle effects imply that stock, real estate, and consumer prices will decline into ’16-’19, and as late as ’21-’22 (secondary inflation- or deflation-adjusted low for prices), along with declines in bank loans, profits, employment/payrolls, consumer spending, private investment, private GDP (and an outright decline in private per-capita GDP), and gov’t receipts.
The US has built out a private and public infrastructure based on perpetual nominal GDP growth of 6-7% with abundant supplies of cheap oil; however, the VERY long-term sustainable rate of economic growth is no more than population plus its replacement in human, physical, and natural capital terms, which implies that the long-term sustainable rate of growth for the US economy is closer to 0% for the next decade or so and no faster than 1-1.25% thereafter. We simply cannot support growth of 7-8%+ gov’t spending AND private debt and debt service from an organic growth perspective.
Local, state, and federal gov’t spending, including public employee and retiree wages/salaries, benefits, and pension payouts, must be reduced by the proportional amount that the private sector grows (or doesn’t) versus the built-in expectation of 6-7% nominal GDP and 7-8% gov’t borrowing and spending.
We can choose to make the cuts and restructure the private and public sectors to adapt to the inevitability of a lower material standard of consumption hereafter but perhaps with little or no decline in well-being; or Nature and the law of exponential mathematics will do it for us. I see no sign that the former is being considered, and by definition Keyensians and debt-based, growth-forever advocates cannot even conceive of such a notion.
Menzie,
The only way you can say we “prevented GDII” is to witness a self-sustaining recovery. My point is that small ball stimulus did not produce one, nor did you expect it to (in that you argued it was too small). Further, by allowing fiscal contraction in a still-fragile economy, the stimulus risks doing to the economy what the first time homebuyer credit did to the housing market. As a result, we risk going right back where we started, except we will have 1) many more long term unemployed; and 2) we will have almost zero chance of “large-ball” stimulus to fix the problem; and 3) we would have wasted precious time in implementing meaningful, painful, inevitable restructuring of the banking sector.
Word is that Zhou-en-Lai, when asked whether he thought the French Revolution was a success, said, “its too early to tell.” I think the same is true of whether the too small stimulus was better than nothing.
CoRev Recovery.gov shows both funds awarded and funds paid out. Contracts, Loans & Grants awarded is only shown at the end of each quarter. As of 30 June 2010 it was $219B. Funds paid out is shown as $145B as of now. Recovery.gov has a link that explains the difference between the two. Roughly speaking, funds paid out reflects actual cash paid out of the US Treasury. Funds awarded refers to monies that have been made available (“obligated” in government speak) to recipients. Economic activity is mainly driven by obligated funds rather than paid out funds. For example, highway monies are not actually paid out by the US Treasury until some fraction of the work has been completed (often 100%). The economic activity begins when construction crews start hiring, buying raw materials, renting machinery, arranging bank loans with the govt contract as “collateral,” etc. That’s the difference.
Fiscal stimulus is not something that you can just push out for one or two quarters, see a jump in GDP and then withdraw. That was the lesson of 1937. That was the lesson of Japan in the 90s. In order to make it sustaining you have to keep pushing out stimulus until it begins to bind (or crowd out) private sector investment.
” We did get a stimulative effect early on, but it was too short to maintain economic growth, and that is why I complain about it being back loaded.’
Not following you here. I agree that it was too short…lots of folks predicted that the recovery would peter out in mid-2010 because the stimulus was too small. But this is an argument for more backloading, not less. What we needed was a stimulus that was ~$1.3T with larger 2009 and more 2010 dollars.
Note that the Recovery.gov site shows $228B in tax benefits to date; however, keep in mind that roughly $70B of that was just an extension of the AMT and cannot meaningfully be called stimulus. It was just continuing existing policy. Withdrawing the AMT would have been contractionary, but perpetuating is not additional stimulus. So total tax benefits to date are more like $158B.
David Pearson: I think you have confused Anonymous (9:27AM) with me.
2slugs, I presume you wrote your first paragraph for someone other than me.
Menzie, 2slugs, is there a study that compares size and effectiveness of stimulus packages? If not, it is sorely needed. I did find this recent WaPo article which compares the G20 for this round of stimuli: http://voices.washingtonpost.com/ezra-klein/2010/06/research_desk_responds_how_do.html .
Coincidentally, in that article they link to IMF’s April data, and it indicates that the net effect of the US stimulus is only 0.66% for 2009 and 10. With the latest data it must approach zero. Here’s the link: http://www.imf.org/external/pubs/ft/weo/2010/01/weodata/weorept.aspx?pr.x=61&pr.y=1&sy=2008&ey=2010&scsm=1&ssd=1&sort=country&ds=.&br=1&c=213%2C193%2C273%2C223%2C156%2C924%2C922%2C456%2C199%2C184%2C132%2C134%2C534%2C536%2C186%2C136%2C112%2C158%2C111%2C542&s=NGDP_RPCH&grp=0&a=
David Pearson Sorry, “Anonymous” was my posting. Something got cleared out in the Name block.
I’m still not following your argument. It’s hard to see how the worst case scenario of adopting a stimulus could be any worse than the best case scenario of doing nothing. If we did nothing, then wouldn’t we still have a problem with long term unemployment? Except that they would have been unemployed even longer. As to your second argument, if I’m understanding you correctly you are claiming that our best bet for a good stimulus package would be to wait until we had an absolute catastrophe on our hands before pushing for a stimulus package. Sorry, but that’s crazy. And in all likelihood it would almost guarantee a Speaker Boehner, Majority Leader McConnell and President Palin. We may have to settle for a Speaker Boehner, but things could be worse and would have been worse without ARRA. Your third argument smacks of Secretary Mellon “Liquidate, liquidate, liquidate” thinking. As though we must all undergo some kind of collective purgative to cleanse the economic system of rottenness. I’m not a big believer in self-flagellation and unnecessary pain. A bit too medieval for my tastes. We don’t need to fix every structural problem with the economy as a precondition to addressing the immediate problem of weak aggregate demand.
CoRev Here’s an ungated version: “Measuring the output responses to fiscal policy.”
http://www.voxeu.org/index.php?q=node/5462
Or “Government Investment and Fiscal Stimulus in the Short and Long Runs”
http://www.princeton.edu/economics/seminar-schedule-by-prog/macro-s09/monetary-fiscal-policy-co/schedule/additional-participant-pa/GovtInvest.pdf
Not only do we face another cyclical recession and bear market period during an ongoing secular bear market (lasting into ’16-’19 to ’21-’22) and debt-deflationary depression, we must prepare for a last-man-standing, Cold War-like contest with China-Asia for global peak oil production hereafter.
Politicians, CEOs, the media, Wall St., and e-CON-omists are not telling you what you need to know; thus, you are very likely not preparing properly for an epochal downshift in your standard of material consumption and longer-term material and psycho-emotional well-being.
Wake up! If you are among the bottom 80-90% of US households, you must prepare for the next 10-20 years during which you will have to adapt to living on one-third to eventually one-half less than what you live on now, and on what you have come to perceive as “normal” or a permanent way of life; it ain’t normal, and it ain’t permanent.
Re: “aggregate hours worked in the private sector continues to rise faster than private sector employment”
This appears to be a major factor in continuing high unemployment. Employers are opting for longer hours for existing employees over hiring new ones.
My son is working 60+ hours per week. He’s up at 2 a.m. every morning, working remotely at home at his computer, to handle tasks that should be done by a night time staff. His supervisor recently asked him to go to 12 hour days, 7 days a week. He told her that was a deal killer and his wife would divorce him.
My brother has been working straight 7 day weeks in an axle factory for months. The plant has been hiring 50 new workers a week, but with a $10 per hour starting salary, the turnover rate has been 50%.