Last May, Governor Romney stated that in a typical recovery, monthly employment increases should be about 500,000 per month [1]. The sheer implausibility of that statement (assessed in this post) has induced him to reduce his estimate (without explanation of the change) to 250,000 per month. [2]. In Figure 1, I provide a plot of the implied path, as well as that from his May statement (which made me laugh for days!). In other words, his forecast has moved from clearly “Heritage Foundation space” to something that seems a bit less implausible, even if not clearly motivated by a specific model
Figure 1: Nonfarm payroll employment, seasonally adjusted (blue), Romney trend assuming 250,000 increase per month, per August 2012 estimate (red), and 500,000 increase per month, per May 2012 estimate (green). Assumes 100,000 increase per month from August 2012 to January 2013). NBER defined recession dates shaded gray. Dashed lines at 2001M01 and 2009M01. Blue numbers denote average employment growth per month between trough and peak. Black number between vertical dashed lines is average employment growth between inauguration months. Source: BLS, NBER, and author’s calculations.
Interesting observations:
- The 500,000 number clearly exceeds that recorded in recent history. Kudos to Governor Romney for disposing of that number.
- While 250,000 jobs per month is more in line with recent “jobless recoveries” (i.e., after the recessions of 1990-91, 2001, and 2007-09), it is still substantially above that recorded during the G.W. Bush recovery (250K >> 91.7K). It is also very much above (!!!) the 11,100 jobs per month recorded during the entire G.W. Bush administration, when we last implemented tax cuts advocated by supply-side advocates.
- 250,000 is also above that recorded in the 1991-2001 recovery (196,700/mo, mostly spanning the Clinton Administrations).
- To my knowledge, the 250,000 figure is not based on simulations from a model. Rather (inferring from the Romney white paper), it is based on extrapolations from two previous recessions, specifically the 1974-75 and 1981-82 recessions.
I recommend the “The Romney Program for Economic Recovery, Growth, and Jobs” for reading, if you want to see how these, and other, numbers were generated. Readers will note that the authors mention, but do not condition on, balance sheet effects, in tabulating the expected growth rates in output and employment. In this respect, they have ignored cross country studies on the issue. [3] [4]
I am dubious that predicted job growth can be sustained at posited rates, if thus-far-unspecified tax expenditures are eliminated, the deficit is to remain at specified levels, and spending cuts along the lines of the Ryan plan, which Governor Romney has endorsed, are implemented. [5]
Update, 8pm Pacific: IMF confirms the findings that growth is typically slower after a banking crisis/recession, in its just released adjunct to Article IV report on the US. [6]
Update, 8/7, 12:36PM: Brad DeLong dissects the white paper, so you don’t have to.
It will be a golden age. A chicken in every pot. Flat screen TV in every room. Freedom from fear. Freedom from want.
11,100 jobs per month recorded during the entire G.W. Bush administration
They were all govt jobs too. 1.7 million govt jobs and -650K private sector jobs.
The average population during the Reagan years was 237 million. Today the population is 309 million. The average number of jobs added in the best month of each of the Reagan years was about 0.17%. Reagan’s best month was September 1983 when 1,114,000 jobs were created, but that was the only month when job creatin exceeded 500,000.
It is interesting to note that most economists believed there would be a serious recession after WWII. Joseph Schumpeter was almost alone in forecasting that pent-up demand would create an economic boom. Schumpeter was right while the others were wrong.
When he is elected and if Romney is successful in reducing the huge wedges to economic growth such as Sarbanes-Oxley and Dodd-Frank and the tax code, while it is a tall order, he could equal the Reagan rate of job creation (Truman after WWII was even higher). Based on the current population and the satisfaction of the pent-up demand, we could see months of over 500K jobs created in the Romney administration.
What is the point of this post? Politicians make things up? Shocking!!!
This post reminds me of the scene in “The Blues Brothers” in which Joliet Jake explains to Carrie Fisher why he jilted her:
Carrie Fisher: “You miserable slug! You think you can talk your way out of this? You betrayed me.”
Joliet Jake: “No, I didn’t. Honest… I ran out of gas. I… I had a flat tire. I didn’t have enough money for cab fare. My tux didn’t come back from the cleaners. An old friend came in from out of town. Someone stole my car. There was an earthquake. A terrible flood. Locusts! IT WASN’T MY FAULT, I SWEAR TO GOD!”
The defenders of the Administration need to talk their way out of a very poor record, especially when Romney is coming along saying that he can do better.
“My clunker ran out of gas before I could get the cash. My stimulus didn’t have enough money. Nobody can produce that many jobs. It was an economic earthquake. A balance sheet recession! Look at the cross sectional evidence! IT WASN’T MY FAULT, I SWEAR TO GOD!”
Ok, OK, let’s look at the cross sectional evidence if we must. Menzie mentioned a new IMF paper that. Let’s see.
Wow, maybe there is something here: there are 86 banking crises studied in the IMF paper. That’s a lot of cross sectional evidence. Oh wait–better not get so excited–many of these are emerging market crisis with multiple financial crises–countries like Brazil, Chad, Senegal, etc. Might be better to just look at non-emerging market cases. We should still have a rich data set.
Uh oh–there are only 13 non-emerging market cases. Not to worry, should be many pertinent examples.
Let’s see what have. The US in 2007 is one data point. We’d better throw that one out. Can’t explain the current climate in the US
by using the current climate as a data point.
Looks like Ireland, Iceland, and the UK in 2008 are also included as data points. But Iceland and Ireland imploded for their own special
reasons and the UK is in double dip, no doubt encouraged by the policy of cutting spending and raising taxes. These don’t seem particularly
relevant and they are all drawn from the great recession period anyway.
What else do we have in the remaining 9 cases? Well, there’s the S&L crisis in the US in 1988. But that wasn’t particularly severe.
What else? I see Finland, Sweden, and Norway in 1991. Those cases might be relevant to the US today since Sweden had a serious banking
crisis at the time. But I seem to remember that there was something else going on. What was it? Oh, yeah, that’s right. Sweden was in the ERM, the precursor to monetary union today. Inflation was accelerating in Sweden over the latter part of the 1980s and Sweden was forced to continually devalue.
In fact, the central bank raised the overnight rate to 12 percent in July 1992 and then to 16 percent in August. During the speculative
attack in September of 1992, the central bank attempted to defend the Krona peg by raising the overnight rate to 75% and then finally to 500%.
That severe monetary policy killed a banking system that was already in crisis and no doubt contributed greatly to the ensuing poor
economic performance. Finland and Norway, since they are tightly integrated to Sweden, came along for the ride.
If we throw those out, we only have 5 cases left. Korea in 1997 is one of the cases? That was part of the Asian emerging market crisis.
That leaves 4 cases–Israel in 1977, Japan in 1997, Spain in 1977, and the UK in 1973. The aftermath of the banking problem in the UK in 1973 happened at the same time as the oil price shock, the world wide stock market crash, and the world wide severe recession. Japan had the zombie banking problem while the US has cleaned up its banking system, etc. etc. I could go on and on pointing out all the differences between the US today and these historical cases. The so-called cross-sectional evidence is just an attempt to find an excuse for the consequences of poor policies of this Administration.
Romney has not really changed his view. When he said 500K, he was talking about what you might see for some months at the beginning of a robust expansion. His advisors have put out a plan that averages 250K per month over his 4 year term. But that won’t be a uniform 250K. In the last year, it’s reasonable to think that growth will be better than population growth, perhaps 175K per month. In the first year, since Romney’s policies will have not had time to take effect, we should see the continuing tepid Obama expansion of maybe 125K per month. That means we have 75K X 12 + 125K X 12 to distribute on top of the average 250K over that middle 2 year period. If the distribution is front loaded, we could easily see some 500K months.
As I’ve said before I think this goal is ambitious. But it’s good to see our future President answer JDH’s question from a previous post, “Is This As Good As It Gets?” with a resounding “NO.” Meanwhile the administration’s defenders are attempting to lower expectations by talking about cross sectional evidence.
Rick Stryker: OK, never again shall I do a cross section analysis, as all individuals are different…. But seriously, what have Romney’s advisors done? They looked to the US only, selected two recessions, neither of which are balance-sheet induced. I can understand why you defend that approach given your views on applied econometrics.
As for your full-throated defense of Governor Romney’s 500K/mo statement, all I can find is this unconditional statement [link], quote: ““We should be seeing numbers in the 500,000 jobs created per month. This is way, way, way off from what should happen in a normal recovery.” I don’t see anywhere any of the nuances you cite. Perhaps you can provide the link to the transcript which includes all those caveats you list.
Please give my regards to the rest of the campaign team.
Rick Stryker When he said 500K, he was talking about what you might see for some months at the beginning of a robust expansion.
This isn’t the beginnning of the recovery, so why did he phrase it using the present subjunctive? Also, where did he get this 500K number from? I went all the way back to 1939 and I was only able to find 16 months that had nonfarm job increases of 500K or more…and one of them was in May 2010!!! So that’s 16 hits out of 882 observations…and somehow that’s supposed to represent a typical recovery??? Wow. You’re really into this 1% thing, aren’t you.
No wonder you like Romney.
Menzie,
Yes, I would agree that the 74 and 80-82 recessions were different from the situation today. The 74,80-82, and great recessions were preceded by an oil shock but the 74 and 80-82 events also had very tight monetary policy while the current recession was exacerbated by a financial crisis coupled with very loose monetary policy. These past recoveries were helped by a recovery in housing which we don’t see today. So, there are some important differences. Personally, I don’t see the 74 recession and recovery as very relevant for today’s events. But I do think the 83-84 recovery is relevant.
I’m not suggesting that cross sectional analysis is useless. Rather it’s just really hard to do since you have to account for all the differences between events you are comparing.
Regarding the Romney quote, I don’t have a transcript. I can tell you what I remember about the interview. The Fox News analyst pointed to a drop in the unemployment rate, which she implied was good, and the jobs report, which she implied was bad since it came in below expectations of something like 160K. She wanted to know which half of the glass Romney thought we should look at. Romney argued that both were bad. First, we should be seeing 500K jobs not 160K. And he pointed out that the unemployment rate dropped because people left the labor force. He also said that the public is wondering why it’s taking so long for the recovery to get started. Instead of getting started, it seems to be getting worse. It was that context, that 500K is what we should see if the recovery had really arrived, that leads me to think that that’s what Romney meant. He wasn’t saying that we should see 500K on a permanent basis.
2slugbaits,
See my recollection of what Romney actually said in my reply to Menzie. I don’t believe Romney used the present subjunctive.
Regarding counting the number of times that 500K jobs happened historically, I’ve already pointed out the fallacy in doing that. As you know, the labor force grows and it was much smaller in the past. You need to scale the 500K by the current size of the labor force and then ask how many times that percentage happened in the past. If you do that, you’ll find that 500K growth (and larger) was much more common than a simple count would suggest.
Rick Stryker I do think the 83-84 recovery is relevant.
Huh? That recovery started off with very high nominal interest rates, which gave the Fed a lot of room to work with. The 81-82 recession was engineered by the Fed and it was within the Fed’s power to end it almost at will. The 81-82 recession had a completely different origin. And it wasn’t as global as this one. And of course there was a lot of new fiscal stimulus. Remember, going into that recession the budget was almost balanced, so a large deficit had a lot of shock potential.
The Romney quote was just that…a quote. I saw him say it. And I’m sorry, but there is no way that you can adjust the workforce numbers to make a 500K per month job growth number plausible. It’s absurd on the face of it. We have never had job growth that strong even after scaling. It might be possible to have one or two outlier months near 500K (and May 2010 was just such a month), but Romney wasn’t talking about outliers. Romney was talking about a rate. And now even Romney has backed off from that crazy claim. A 250K rate is at least on the edge of plausibility, although Romney hasn’t offered up any plausible plan that would have a snowball’s chance in hell of achieving such a target.
Rick Stryker: Video of quote here. The quote is accurate. You know, having commented on this post, how rare 500,000/mo in percent terms is over the past quarter century.
Thanks for the video link Menzie. I think it does confirm what I remembered, that the context of Romney’s remark was that we are still waiting
for the recovery and when it happens we should be expecting 500K jobs, not 160K.
I know that we haven’t seen 500K numbers in the last 25 years. But that’s not very surprising at all. During that time, we have experienced two
fairly mild recessions in which the unemployment rate rose gradually and then slowly came down, the so-called “jobless recoveries.” And we’ve also had a huge recession with a recovery that’s failed to materialize. We shouldn’t expect to see gains of the magnitude of 500K over a short time
after the two mild recessions, since we didn’t see big losses either. On a scaled basis (scaled to size of labor force), we saw an average 149K loss with the largest monthly loss equal to 370K in 1991. From February 2001 to July 2003, the average monthly loss on a scaled basis was 92K with the largest being 336K. I don’t think the “Great Moderation” is the proper period to evaluate what we should expect from the current recovery, which followed a very violent recession.
If we go back to the 1980-2 recessions and before, we see recessions that look much more like what we’ve just experienced, with significant drops in output and rapid increases in the unemployment rate. Below, I’ve listed out each recession since the late 1940s and reported the monthly employment gains (in thousands) in the subsequent recovery, scaled to the size of the labor force at the time.
Recession Dates
1948-11-01, 1949-10-01 Job Growth
10/1/1949 906
11/1/1949 833
12/1/1949 43
1/1/1950 -714
2/1/1950 1,983
3/1/1950 1,273
4/1/1950 1,016
5/1/1950 1,085
6/1/1950 1,082
7/1/1950 2,118
8/1/1950 732
9/1/1950 770
10/1/1950 188
11/1/1950 219
12/1/1950 1,223
1/1/1951 807
2/1/1951 818
Recession Dates
1953-07-01, 1954-05-01 Job Growth
10/1/1954 637
11/1/1954 411
12/1/1954 447
1/1/1955 395
2/1/1955 851
3/1/1955 750
4/1/1955 702
5/1/1955 729
6/1/1955 510
7/1/1955 331
8/1/1955 390
9/1/1955 438
10/1/1955 416
11/1/1955 548
Recession Dates
1957-08-01, 1958-04-01 Job Growth
12/1/1958 998
1/1/1959 521
2/1/1959 827
3/1/1959 760
4/1/1959 570
5/1/1959 320
6/1/1959 310
7/1/1959 -1,164
8/1/1959 227
9/1/1959 -172
10/1/1959 686
11/1/1959 1,328
Recession Dates
1960-04-01, 1961-02-01 Job Growth
1/1/1962 715
2/1/1962 215
3/1/1962 779
4/1/1962 60
5/1/1962 43
6/1/1962 244
7/1/1962 220
8/1/1962 331
9/1/1962 152
10/1/1962 33
11/1/1962 -67
12/1/1962 211
1/1/1963 273
2/1/1963 215
3/1/1963 608
Recession Dates
1969-12-01, 1970-11-01 Job Growth
11/1/1971 488
12/1/1971 620
1/1/1972 380
2/1/1972 535
3/1/1972 397
4/1/1972 551
5/1/1972 529
6/1/1972 -92
7/1/1972 769
8/1/1972 235
9/1/1972 721
10/1/1972 521
11/1/1972 540
12/1/1972 617
1/1/1973 696
2/1/1973 470
Recession Date
1973-11-01, 1975-03-01 Job Growth
6/1/1975 432
7/1/1975 667
8/1/1975 135
9/1/1975 521
10/1/1975 247
11/1/1975 577
12/1/1975 830
1/1/1976 526
2/1/1976 391
3/1/1976 410
4/1/1976 30
5/1/1976 109
6/1/1976 285
7/1/1976 264
8/1/1976 314
9/1/1976 22
10/1/1976 551
Recession Date
1980-01-01, 1980-07-01
1981-07-01, 1982-11-01 Job Growth
3/1/1983 412
4/1/1983 412
5/1/1983 560
6/1/1983 616
7/1/1983 -455
8/1/1983 1,627
9/1/1983 395
10/1/1983 511
11/1/1983 514
12/1/1983 643
1/1/1984 685
2/1/1984 392
3/1/1984 516
4/1/1984 436
5/1/1984 535
6/1/1984 439
7/1/1984 338
8/1/1984 435
9/1/1984 399
10/1/1984 485
As you can see recovery months were often greater than and in some cases much greater than 500K. Here are the monthly job losses during the
most recent recession, scaled to current employment levels:
1/1/2008 -81
2/1/2008 -92
3/1/2008 -201
4/1/2008 -184
5/1/2008 -192
6/1/2008 -204
7/1/2008 -267
8/1/2008 -422
9/1/2008 -480
10/1/2008 -792
11/1/2008 -655
12/1/2008 -816
1/1/2009 -726
2/1/2009 -806
3/1/2009 -702
4/1/2009 -367
5/1/2009 -492
6/1/2009 -347
7/1/2009 -237
8/1/2009 -204
9/1/2009 -208
10/1/2009 -43
11/1/2009 -176
12/1/2009 -41
1/1/2010 -36
Those losses are quite big, just as they were in recessions before 1990. Romney and his advisers are asking why we don’t see job increases of equivalent magnitude in the recovery. Not every month of course, but we should some big recovery months. Don’t you think that’s a reasonable expectation?
2slugbaits,
In the great debate between rules and discretion, discretion won before the early 80s, with constant discretionary fiscal and monetary policies, unpredictable rules, and expansion of government. The early 1980s represented a fundamental transition in policy to rules-based policy and less government intervention. The recent period has been characterized by a return to discretionary policy: stimulus, the unprecedented monetary policy actions, implementation Basel 2.5 and Basel III, Dodd- Frank, regulations, and Obamacare. Romney is proposing a return to a more rules-based policy with smaller government, lower taxes, and repeal and replace of these programs–in short a return to the general policy stance that began in the early 80s. That’s why I look to that time as relevant for today.
I think this Romney is a cult-believing nuthead and
a man of the war machinery. This ultra-rich man does not care about employment and its statistics.
My bet is that he will become president. Stocks doing good in wartime will rise.