Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
A Rescue Plan for America’s Heartland That’s About Work, Not Cutting Checks
Via: The New York Times.
What does America look like to an economist? In the new working paper “Jobs for the Heartland: Place-Based Policies in 21st Century America,” Benjamin Austin, Edward Glaeser, and Lawrence Summers neatly outline the problem of “left behind” places in the US economy (bold by me):
We divide the U.S. into three regions: the prosperous coasts, the western heartland and the eastern heartland, divided based on year of statehood. The coasts have high incomes, but the western heartland also benefits from natural resources and high levels of historical education. America’s social problems, including non-employment, disability, opioid-related deaths and rising mortality, are concentrated in America’s eastern heartland, states from Mississippi to Michigan, generally east of the Mississippi and not on the Atlantic coast. The income and employment gaps between the three regions are not converging, but instead seem to be hardening into semi-permanent examples of economic hysteresis.
As the economists note in today’s follow-up New York Times piece, the relative GDP of what they call the “eastern heartland” region would be 50% higher had it grown at the rate of America’s coastal states for the past half century, and more than double had it grown at the rate of the western heartland states.
So next steps? Simple enough. Focus ameliorative policy on the troubled region. But that’s not typically how we do things in the US versus, say, Europe. Policy tries to help poor people rather than poor places. In the European Union, it is just the opposite. And it is along those lines that Austin, Glaeser, and Summers dismiss some commonly proposed place-based ideas such as directing federal activity (there just are not that many federal workers) or infrastructure spending (such projects have a mixed record at encouraging local employment) toward the troubled region.
A better idea, the authors write in the NYT, is for Washington to create a regionally-targeted wage subsidy that would boost workers’ hourly wages, “which would effectively increase the national minimum wage without discouraging employers from creating new jobs.” (When Glaeser has written previously about the wage subsidy idea, he’s suggested a $3 per hour boost.) The main downside highlighted is the cost of the subsidy. On that issue:
The wage subsidy can be lowered in labor markets with low joblessness, and it should be phased out entirely for the prosperous and the long-term employed. The subsidy should be highest for workers who have been jobless for a long period, and for veterans, or for those overcoming a labor market disadvantage.
(I have previously blogged about the general idea here and wrote a New York Times column on it here.)
And on that cost issue, I would refer you to this essay by Glenn Hubbard on novel labor policies, in which he approaches the cost issue differently:
As my Columbia colleague and Nobel laureate Edmund Phelps has observed, the loss of opportunities for work implies more than just lost wages for directly affected individuals. Such loss leads to negative externalities in crime, drug use, high tax rates to fund social assistance, and other ills. Phelps argues that funds to support work could be drawn from amounts saved by reducing negative externalities through low-wage subsidies for employment.
One more thing: Note how the authors — including Summers, a prominent Democratic economist — push back against the idea of a universal basic income as a possible solution:
The fans of programs that accept, and even encourage, joblessness, like universal basic income, seem to forget that human satisfaction doesn’t come primarily from material comfort, but from purpose, a feeling of accomplishment and the social support that often occurs in a work environment. An America in which 40 or 50 percent of adults live without working, relying on the generosity of a federal handout, is a nightmare.
Previous I have pointed out how at least one potential 2020 Democratic presidential candidate has nixed the UBI.
Published in Economics
You mean like how Obama nixed the individual mandate when running against Clinton?
Let’s say the government massively increases fiscal transfers to poorer regions. Price levels would rise, as would U.S. inflation as a whole since, let’s face it, deflationary pressure in the heartland is why coastal inflation doesn’t show up in national statistics. So interest rates go up for everyone. Meanwhile, the main competitive advantage of “the heartland” vis-a-vis the coasts will be taken away, i.e. the lower cost of living.
This is pretty basic macro. There is an argument to be made for wage subsidies (I’m agnostic on the question), but regionally targeting them would be insane, not unless we’re talking really small geographic areas.
At the risk of being the skunk at the garden party, if you are living in an area with high unemployment and lower income, and there are jobs available in a different part of the country, why not move to where the jobs are?
I choose to live in a county where there is a high cost of living and low wages. However, the quality of life is higher than it was in the Phoenix Metro area. We all make choices in life.
The groupings are insanity. What at all is similar about CT and NC?
Even within states, it can be a best of times/worst of times situation.
They probably just don’t know that there are more jobs elsewhere in the country. I had a friend who would travel around looking for work in about a 100 mile or so radius around her hometown during the recession. Going further than that during the recession was too big of a risk.
I must admit, I hate this logic. A few years ago liberals were calling poor white people stupid for not moving to high-income cities where the good-paying jobs were. They were making the extremely elemental mistake of failing to account for regional inflation, which if you do reverses the picture–lowskilled wages are actually much higher in the heartland than on the coasts.
I moved from Michigan (Eastern Heartland) to Texas (Western Heartland) in 1979. My wife and I left our home town, the city where we were both born, and left our families behind, to move from one end of the US to the other.
It was a big gamble, but it paid off. We also moved from the slowest growing region to the fastest. We are still here, and plan to stay. And we should be punished for our courage in making this jump by way of extra taxes to subsidize the poor choices of the sector of the country that are growing most slowly? Why exactly? Seems like something so stupid only a professional economist could come up with it.
If this foolish scheme were implemented, I’d consider supporting Texas independence.
I feel like a broken record, but wouldn’t a better use of money be on subsidizing employers who offer (or invest in) current-tech work training programs for low and mid skill work?
I keep hearing how technology is changing the job outlooks and colleges are too expensive and ineffective at preparing people for work (see Caplan), especially if you are too white and poor to get into a college.
There is a word used here that ought not be allowed: “subsidy”.
When it comes to such a thing, just say no.
How about if we rotate the nation’s capital around, so that other areas could enjoy the largesse of The Swamp. 20 years in St. Louis, then 20 years in Detroit,…
At least let’s send the Department of the Interior to Denver and the Department of Agriculture to Kansas City!
I like Ed Glaeser, and he knows better than this. Those groupings are madness. All of them are approximately like the old joke of averaging out you, me, and Bill Gates. Hey, on average, we’re all rich.
But it’s worse than that. State level aggregation isn’t even reasonable. New York State isn’t rich. New York City is. And for that matter, Upstate New York has a higher per capital SNP than NYC, it’s just that there are a lot of very rich people in NYC and the a whole lot of poor people there.
Counties are the proper measure here, where you see the issue is urbanization/suburbanization -and much though I like Glaeser, he has never grasped that decentralized suburbanization is how Americans get rich. People don’t need to move to the coasts, they need to move closer to -but not necessarily into -the major cities of their state.
This analysis is horrible.
This analysis strikes me as very misleading, as it does not appear to account for population changes. What is reported in the striking graph is not per capita GDP growth by region, but overall GDP growth.
Well, there have been big population shifts over the period shown in the graph (1965-2016). It does not appear that the authors took this into account at all.
So, I did my own analysis. I’m not sure if this is 100% accurate because, for example, it is not clear whether DC is included in the “Coastal States” figures, and it is not clear whether Alaska and Hawaii are included at all. I left DC, Alaska and Hawaii out of the analysis, and calculated the population changes in the three regions reported, from 1965 to 2015. This differs by one year from the period used in the graph shown in the OP, because my source for population information (this Wikipedia page that cites St. Louis Fed data) only goes through 2015. The following population figures are my own calculations in Excel, which have not been double-checked, while the GDP growth rate is from the chart in the OP
I then used the population figures to calculate the changes in per capita GDP by region. This was a bit tricky, but I think that I got it right. Here are my results:
This is far less dramatic than the extreme divergence suggested by the OP, which is largely driven by population changes. I don’t mean to criticize JP for this specifically, as he is just copying someone else’s chart. However, I submit that this shows that the huge gaps in GDP growth suggested by the chart in the OP are misleading and sensationalized. This is mostly about population shifts.
The really troubling thing is that it took me, oh, about 20 seconds of thought to realize that the chart was probably misleading because of significant population shifts. It then took me about, say, 20-30 minutes to do the calculation above.
I cannot access the NBER paper that is cited in the NYT article referenced in the OP. It is called “Jobs for the Heartland: Place-Based Policies in 21st Century America,” by Benjamin A. Austin, Edward L. Glaeser, and Lawrence H. Summers. These are the same authors as the NYT article that has the misleading graph.
[Continued]
[Continued]
Now, if my analysis about the misleading nature of the GDP growth graph is correct, I submit that it is extremely troubling that the NY Times would make a mistake like this, and even more troubling if the “Lawrence H. Summers” who is an author of both the NBER paper and the NYT article is who I think he is — the famous Larry Summers. And I’m reasonably confident that he is, as the NBER article does allow me to link to his affiliation (the Harvard Kennedy School of Government), which in turn lists him as a professor and details his career accomplishments.
Larry Summers was the Secretary of the Treasury in the Clinton Administration; then the President of Harvard; then the Director of the National Economic Council in the Obama administration. He is a hugely influential public figure. Despite the fact that I am generally on the opposite side politically, I generally consider Mr. Summers to be a man of tremendous talent and substantial integrity.
I find it very disappointing that a public figure and economist of Mr. Summers’ stature would use such a misleading graphic in an article in the New York Times.
This graph seems misleading, the “late 70’s” seems an arbitrary time to start. In this 40 year time span the Midwest has seen the rollback of the US auto industry and the rise of liberal leaning state governments. A graph of the 40 year time span prior to this (late 30’s up to the late 70’s) would look MUCH different. Times change, regional economies have to adjust.
If we line up all Federal and State income transfers and history of heavy industry manufacturing and industrial unions we’d see correlation with the eastern heartland. But it doesn’t matter, instead of grouping these states which hides causes, we should desegregate all the states and look for commonalities among low performers. Even so, if our response to everything remains more money for the poor we’ll never get at the causes. Low income is a symptom of cultural characteristics and bad government and always has been everywhere.
Instead of transferring more money perhaps we should eliminate income transfers, or eliminate minimum wages so on the job training, which is how we learn to work, pulls people into the work force and pulls investors to non minimum wage locations. We could replace public schools as we know them with school and training choices. Or explore other public goods and regulatory issues that may be having negative effects but we can’t seem to get off the same old response. We’ve now had global experiments for nearly a century in historical materialism and we should know by now that it’s the cause not the solution.
I live in the “poor” Eastern Heartland. For the record, we are doing just fine. Every where I go I see help wanted signs. So there are jobs available, and wages are going up. My company recently gave every hourly employee a $1 / hr raise, just because of job market pressures, and we already paid pretty good. Homes are affordable. I have less than $300,000 dollars invested in my 2 year old, 4000 square foot home on 24 acres, less than 10 minutes from work.
Regardless of what the economic data says, I think we have a much better standard of living than either of the coastal areas.
That is actually what has happened. Most that could’ve, have left. The area in question has largely seen its best from each generation leave, since the 1950’s. The area has a much larger percentage of folks working in rural industry sectors such as extraction (mining), agriculture, forestry and tourism related to those sectors. The population in these areas is older and has far higher levels of physical/mental handicap. I work with the Appalachian Regional Commission through my job. It has several programs similar to what the author is proposing. They have all been failures.
For me, the bottom line is this, the area in question just isn’t going to have the growth and life-style of the other areas. That is not the reason for the anti-social issues that it is facing. Much like the inner-city, these areas have massive resources. These resources are just being extracted out in a way that doesn’t accurately compensate the people who provide those resources. Government action has created this disparity. Now, we are being told that the same people that are causing the problem are the ones that should fix it, by doing pretty much the same thing that caused the problem in the first place.
The paper appears to be available for free here. The title is different, but the abstract is 100% identical and it contains the exact same graph as the NYT article.
From skimming the paper and the figures, it is clear that the authors did not make the “mistake” you suspect. They acknowledge population migration, discuss per capita growth (at least a little), and most importantly, show this graph right below the one copied in the NYT:
Granted this is at the level of “per worker” and not “per capita”, but it seems to reflect your calculations well.
The take-home message is that the authors didn’t make rookie mistakes. They just wrote a very staid (and not-at-all novel) economic paper, and then penned a NYT editorial on top of it that sounds like somebody auditioning to be Krugman’s replacement when he strokes out. Indeed, the graph they show in the NYT article is hardly mentioned in the paper itself.
The irony is that the core message of the paper is one conservatives can actually find some common ground on:
That’s not too far off from the principles of federalism and subsidiarity.
But sadly, yet predictably, the authors felt it necessary to package the whole thing into a “how can we help those flyover rubes?” context.
Mendel, thanks for this. I really didn’t expect the authors to have overlooked the obvious population-based explanation. Unfortunately, this means that they were being consciously misleading, as I suspected.
I notice that this GDP per worker graph overemphasizes the divergence by having the y-axis cross at around $40,000, rather than zero. This is sometimes acceptable, because where a change is small, using the full dependent variable range may make it difficult to see the differences. But it requires a more sophisticated reader (which I would expect in a Brookings or NBER paper).