Perry’s employment record in Texas

By Felix Salmon
August 15, 2011
Paul Krugman has an important column today about Rick Perry's record of job creation in Texas:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Paul Krugman has an important column today about Rick Perry’s record of job creation in Texas:

So where does the notion of a Texas miracle come from? Mainly from widespread misunderstanding of the economic effects of population growth…

Texas tends, in good years and bad, to have higher job growth than the rest of America. But it needs lots of new jobs just to keep up with its rising population — and as those unemployment comparisons show, recent employment growth has fallen well short of what’s needed.

The unemployment numbers are interesting, but I thought it would be interesting to look at the employment numbers instead — and to see how employment in Texas compares to employment in the rest of the country. So Nick Rizzo collated the data for this chart, taking employment figures from Google’s Public Data Explorer, and filling it out with population data from the Census Bureau and — for the 2011 population of Texas — the Texas Department of State Health Services. Here’s the result:


The employment-to-population ratio in this chart is lower than the employment-to-population ratio we normally see, because it includes everyone, from infants to convicted felons. According to the figures we have for 2011, 44.7% of the total US population has a job, compared to 43.5% of the Texas population.

And Perry’s record is pretty bad, here: he inherited a ratio of more than 47% in Texas from George W Bush, and has presided over a steady decline ever since — including every year of the Bush presidency bar 2005.

The single most important task facing the US is to turn the employment numbers around and get the employment-to-population ratio rising again. Obama has been bad on this front. But Perry’s decade-long record in Texas is no better.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

Nice chart. Doesn’t the steeper decline in Texas seem to imply that the state has become a retiree haven?

Posted by RZ0 | Report as abusive

I think you need to make a bit more of a case that employment ratio is a better metric than unemployment rate. If Texas has more children or retirees or wealthy housewives than the rest of the country why is that relevant?

Posted by right | Report as abusive

Thinking about this more, I also wonder if illegal immigrants are treated the same way in both the employment and population data sets.

Posted by right | Report as abusive

Census has the population breakdown, age group as a % of total population for Texas:
>18: 27.3%
20-49: 42.4%
50-64: 17%
64+: 10.3%
>18: 28.2%
20-49: 45.3%
50-64: 13.4%
64+: 9.9%

A few more retirees in 2010 but fewer kids. The prime working age population looks a bit higher in 2010 and wouldnt explain the angle of the Texas line.

Posted by jason11 | Report as abusive

right is right. Texas has a monster illegal immigrant population and many of them just don’t get the kind of jobs that show up in the data.

For goodness sakes, Felix, do you know anything other than partisanship?

Oh, here comes a possible Republican nominee, quick, attack!!

Not only that, but your analysis misses the point completely in terms of what is relevant.

If a state has a low birthrate (Massachusetts) it is easier to have a higher employment ratio. In a demographically vibrant state (40% higher b.r. in Texas), more kids means:
(1) Fewer adults
(2) More stay-at home moms

I wonder what the employment ratio was in the fifties and sixties. Surely it was lower than now. The employment ratio in my 3-child household is 20% and we are doing fine thank you.

On another note, there are social factors in this have nothing to do with the economy.
(1) An increase in stay-an-home parenting as a preferred option
(2) Decrease in intact families means many uncommitted, unmarried men don’t actually feel they need to work. This is a national problem.

Posted by DanHess | Report as abusive

Illegal immigrants isn’t the issue — they’re part of both the population and employment estimates (“foreign born”). In any case, they would also show up in other large states’ data (California, Florida, NY, etc.) as well as in Texas’ numbers.

Posted by BigBend | Report as abusive

Probably need to clarify: not saying ALL foreign-born are illegal immigrants, just that they would be part of that population and employment estimate subgroup, along with legal immigrants, naturalized U.S. citizens, etc.

Posted by BigBend | Report as abusive


Retirement population is not a concern, as the biggest growth in Texas’ population has been between ages 20 and 64:

Posted by GRRR | Report as abusive

About 1/3 of the jobs have been in the government and public sector. About 1/2 of those – meaning 1/6th – have been in schools. That is as the state’s population has grown they’ve added a very large number of government and public sector jobs because people need services.

Would Perry crow about adding 300k public sector jobs?

Posted by jomiku | Report as abusive

right, but if he used unemplyment then he wouldn’t get a graph pointing downwards which is the whole point.

Posted by Danny_Black | Report as abusive

Um, isn’t population growth a good thing?

And I see the lines differently. Texas didn’t move up to unsustainable levels in 2000 and from 03-07, and so it avoided the bust from 07-10. Isn’t that also a good thing?

Fere libenter homines id quod volunt credunt.

Posted by Publius | Report as abusive

Good observation on housewives. That might explain some of the difference between TX and national employment rates, but not the change over time which was the thrust of Perry’s argument. Since housewives aren’t seeking jobs, that paints the opposite picture from F.S.:

@GRRR: what’s the source of your graph? It looks mislabeled to me: you’ve got absolute numbers on the bars and “share of pop” in the title. Check out the percentual change 2000-2010 in jason11′s comment and you’ll come to the opposite conclusion: the population is getting older with fewer kids, 20-50′s moved to 50-64′s, and a few more retirees (in relative terms).

@BigBend: It is reasonable to imagine that illegal labor arrangements are undocumented (unlike foreign-born contracted labor), but that the illegal immigrant will still show up in population statistics. That probably does contribute to upward pressure on unemployment statistics in CA and FL as well. To know for sure, we would need statistics on undocumented labor.

I am curious about the causal relationship P.K. and F.S. seem to be implying. Is population growth causing higher employment (in public sector jobs, as stated) or is immigrant population growing because of (perceived) employment growth? That should be tested.

Posted by dooney | Report as abusive

@Dooney – US Census, 1990, 2000, 2010

Posted by GRRR | Report as abusive

A couple notes:

1) DanHess: Texas having a large number of illegal immigrants working in undocumented jobs does not argue for Texas being a good jobs model. Quite the opposite. Generally in the US, we would like to see legal workers working documented jobs: subject to the minimum wage, paying payroll taxes, etc. “We have lots of illegals working black market jobs” is not a great labor argument for Rick Perry.

2) Dooney: I don’t believe a causal relationship is necessarily implied. The main point is that Texas is adding more jobs than the average state, but has higher population growth than average as well, meaning the unemployment rate is still pretty bad.

3) Publius: pop. growth is not necessarily good or bad. The point is simply that job growth needs to exceed pop. growth to cut into unemployment. It doesn’t in Texas. Your idea that Texas avoided “unsustainable” levels of employment is just odd. (You know it’s not a graph of debt, right?) The graph shows Texas as having a lower employment/population ratio than the US since 1999, up through and past the 2008 meltdown – that’s not a good thing for Texas, it’s a bad thing.

Posted by Dilettante | Report as abusive

I guess all those people moving to Texas from California, Illinois, Florida, Michigan and Ohio must be doing so because of the poor record of employment growth in Texas. es-and-demographics/census/census-data-s how-migration-patterns-texas/

Posted by JohnOmeara | Report as abusive

Uh, no. People moving to Texas are coming from those places because those are the big-population states. Look at the Census data behind the Texas Tribune maps. CA, with almost 37 million residents in 2009 (American Community Survey, also on Census website), saw 546,589 of its residents move to another state — 61,270 to TX). Another 5.8 million Californians moved, but to somewhere else in California. The people that left? Only 1.5% of the state population. TX, with 24.8 million residents, saw 380,932 of them leave for another U.S. state. That’s also 1.5%. Rates of people leaving are about the same; TX Tribune is talking about totals.

Posted by BigBend | Report as abusive

Uh, yes. It is the net migration (moving in – moving out) between states that matters not the percentage leaving each individual state because they all have different denominators. If Texas and California were the only states in the union and each year they lost 1.5% of their population to the other state then more people would leave California for Texas every year until they had equal populations. Would you call that equal migration? Looking at the percentage of people leaving a state without looking at how many are coming in is ridiculous.
Your first sentence gives the primary fact, more people moved to Texas than moved out and this is a zero sum game. The only question is why. I would guess that job opportunities would be a prime reason. Felix seems to think that despite the fact that Texas outpaced the country in job growth (both before and since 2008) from the time Perry took office until now and despite the fact that people are moving there that Texas has been no better at job creation then the rest of the country.

Posted by JohnOmeara | Report as abusive

Uh, still no. Looking at the total # of people leaving a state is not all that helpful, without adjusting for the total # of potential leavers (most of whom stay put or relocate within a state). I do agree that NET migration is the issue; not just domestic, which the TX Tribune maps are about, but also international. Beyond migration, there are also births & deaths; it’s the sum (net) total of all of those that increase a state’s population. The original blog compared population growth to employment. You can’t pick out one component of it and say that it alone explains how Texas is better at job creation. I could just as easily say the 400,000 kids born in TX every year created the need for thousands of public school jobs (govt jobs). Is that the Texas model for job creation? Jobs paid for with taxes?

Posted by BigBend | Report as abusive

Uh, still yes. Looking at the percentage of people of people leaving a state is still useless without looking at the percentage moving in. Again, if Texas and California each traded 1.5% of their population every year then there would be a net migration to Texas for the next 13 years. Exactly how helpful is the fact that they each had the same percentage of ‘leavers’ in that case?
The original blog post didn’t just compare the population growth to employment, it said that it showed Texas has been as poor or worse as the rest of the nation at job creation. I am asking, if that is the case, why is Texas getting an outsized portion of the migrating population?
You could easily say that a net increase of 400,000 kids born in Texas is creating all the jobs in the state but would that have any basis in fact or would you just be making it up? From my reading of the BLS statistics, 75% of the jobs created in Texas since 2001 have been in the private sector but maybe you have a better source for your number. I can point to the fact that (according to the Census Bureau) 130,000 people migrated to Texas from other parts of the country last year. Of all the people I know who have ever moved to another state, job availability is one of the primary reasons why. If you do not think that job opportunity is a major cause of the net migration perhaps you could share your thoughts on why that many people are moving to Texas.

Posted by JohnOmeara | Report as abusive

4 points
First of all, the political/media battle is already lost: in the media, texas has a lot of jobs; media people aren’t reflective, so that is sort of set in stone. Check out the 1st page of todays Boston Globe, an article titled “Romney stands firm amid Perry whirl” quote”despite the strong record of job growth in TX under Perry’s leadership…”
Ok, the battle, in the sense that counts (influencing undecided voters ) is already over.
Remember, a national election isn’t about you and me – it is about the small% of undecided voters who live in swing states. For them, the argument is over, cause they pay attention to newssources that are not gonna deviate from the line.

This post by salmon, and similar posts by krugman, and posts on the other side ( 0)
are lazy and half assed and not very convincing.
I don’t know what the truth is, but it is clear that neither salmon nor krugman nor political math blog have spent anywhere near the amount of time and energy it would take to get an accurate picture – as one commenter on this blog noted, what about stay at home moms, how does that skew the #s vs other states

I work with a guy from texas, he just got back from visiting his mother, and his comment is, its booming donw there – construction is just everywhere.

Lets say salmon is right. How on earth are complex arguments like this gonna win against perry, looking good in a suit on TV, saying: I created more jobs then obama. Does anyone on this blog actually think convoluted graphs are gonna play well (and, for this discussion, the graph is convoluted; I can calculate a std deviation on a regression line, so I’m not talking about me or you – again, I’m talkin about the undecided voters paying attention for 5 seconds to the TV)

Posted by joeenuf | Report as abusive

JohnOmeara: My thoughts on why “that many people [130,000?] are moving to Texas?” My thought is, that’s not that many people, as a share of how many live in other states (.05%), or are already here in TX (0.5%), or compared to how many are born here every year (over 400,000). Previously you said, “…all those people moving to Texas from California, Illinois, Florida, Michigan and Ohio must be doing so because of the poor record of employment growth in Texas.” It’s just not remarkable (to me anyway) that TX gets its “movers” from large-population states, especially when 98.5% of their residents, like TX’s, aren’t leaving at all. (If I were a U.S. university president and my freshman applicants were mostly from CA, TX, NY, IL, etc.–isn’t that what I should expect?) Migration data, personal anecdotes aside, don’t explain all or even most of the job growth. People are moving to TX, yes, and people are leaving TX (would help to know why, or even who they are; families with children? retirees? military spouses?) But that’s only one part of pop. change. The # of children born in Texas is in fact 405,242 (in 2008 – similar to 05, 06, 07), says the TX Dept. of State Health Services – over three times the 130,000 people you cite as moving to TX from other states. The long-term effect of lots of births (& a high birth rate): I’ll go back to ’91 when I find comparable #s, but meanwhile, the Texas Workforce Commission database AreaSelection.asp?tableName=Ces shows that the local govt-education sector in TX gained 39,452 jobs between 2008 Q1 and 2011 Q1. State govt-education [universities] gained 15,348 jobs. The TX private sector lost 203,460 jobs in that same period, of which 109,833 were manufacturing jobs lost. Maybe those IL, MI, and OH workers who came here for factory jobs would like to teach instead? Sorry, looks like they got here just in time for public school job cuts (which other states have been making for a couple years now; the federal stimulus propped up Texas school employment for a couple years, but time’s almost up on that).

Posted by BigBend | Report as abusive

It is not remarkable to you that Texas gets it movers from the large states because you refuse to acknowledge the fact that people are choosing to move to Texas has any importance. Your university analogy only makes sense if you look at it the same way too – people have to apply to college, people have to move to Texas. A real analogy would be to compare universities in different states. If a school in one state was getting lots more applications from residents of every other state than any other school then it might be reasonable to expect there is a reason behind that. And dude, if you want to believe that IL, MI and OH factory workers are moving to Texas to become teachers then go right ahead. Felix’s original post and your responses keep trying to prove that the better employment situation there is due to some freak demographical storm while ignoring the fact that Texas has outperformed the country in every significant employment statistic (public and private workers) during this recession. I don’t know enough about the state to say if Perry did something to make that happen or perhaps it is benign neglect or how big a factor the oil and energy boom has been but attempting to say that Texas has been no better economically than the rest of the country over the last few years is silly.

Posted by JohnOmeara | Report as abusive

People moving to Texas does have SOME importance, but only 1/4 of our growth is explained by that. In the same way not everyone HAS to apply to college, not everyone HAS to move, and very few people do move, relatively speaking. But the media and many others keep harping on migration and immigration, when it’s the high birth rate that’s causing 3/4 of TX growth. (To be fair, that in turn is partly the result of historical migration over the last 150 years, or 500 years, depending on how you look at it.) It’s not a “freak demographical storm” if it goes on for decades, and it’s not anything one governor can take credit or blame for, and it has not successfully been dealt with or our unemployment rate would be lower. Like the nation, Texas has not been able to recover to pre-recession levels of manufacturing or other private-sector jobs — so what “outperformance” do you mean? I went back to 1991 (US Bureau of Econ Analysis website) and local govt jobs (2/3 of which are in schools) grew by 53% between then and 2009 — about 1 in every 8 new jobs in TX. Private-sector grew by 44% (and how many of those jobs are because teachers, coaches, cafeteria workers, etc. had paychecks to spend? Or because schools were built?) The oil and gas booms/busts affect per capita income; TX is near the US avg on that. Our other economic statistics, you won’t hear much about on the campaign trail: poverty, low-wage/no-benefit jobs, illiteracy, income inequality, to name a few. Texas’ showing on these has stunk for decades — but didn’t stop GWB from becoming president, just as Clinton’s Arkansas didn’t sink him. You know what else TX ranks really low on? Voter turnout.

Posted by BigBend | Report as abusive

BigBend, thanks for your insightful work.

On the retirees: We would need to know how rich these immigrants from other states were. It stands to reason that poor seniors would die younger than rich ones.
So, the number of seniors might be the same, but if the group got wealthier, this too, might account for some job growth.

So, the original blogger, FS, point stands:
Perry did worse than the national average in job creation.

Aside from that, Krugmans critique was “Texan experience offers no useful lessons on how to restore national full employment” because “every state can not lure jobs away from every other state”, which is what Texas has been doing. Ya can’t leech upon yourself. Well, you can, but you’d be dead.

Why did Texas do well? Cos of a “strict regulation of mortgage lending.” Also not Perry’s credit.

Finally, Krugman points out to us that providing more jobs for a growing population is not a great trick. An example: you have a population of 500 people, 220 are working. The population grows to 750 people, now 300 are working. Is that good? No. because if jobs had kept up, it should have been 330 jobs, not 300.
It’s even WORSE, if the other states have shrinking populations, yet, they still manage to have some job growth.

Also, Perry used a lot of federal cash in creating jobs like teachers. That wouldn’t be so bad, if he had had the cojones to admit that he took federal money. He doesn’t because he knows that as long as hey SAYS he didn’t, NO ONE will hold that against him. I don’t mean Teabaggers, but all the interviewers from serious media of course, excluding FoKKKs.
If Perry becomes the nominee, it will be because the press hasn’t grilled him on his hypocrisies, his poor decision making in harassing pregnant women instead of focusing on jobs, and his megalomania with his Toll Roads for Texas. Yes, Capitalist Cronyism is a good word for him.

Posted by TheSteelGeneraI | Report as abusive

The chart doesnt take into account the large Hispanic pop which has a higher birthrate and therefor a larger percentage of children who are not in the work force at all but are counted as unemployed for this survey had to look hard to find something to get on perry nice try but no cigar

Posted by Dogman11 | Report as abusive

The chart doesnt take into account the large Hispanic pop which has a higher birthrate and therefor a larger percentage of children who are not in the work force at all but are counted as unemployed for this survey had to look hard to find something to get on perry nice try but no cigar

Posted by Dogman11 | Report as abusive