Targeted? Infrastructure Spending by Unemployment Rate

by on February 18, 2009 at 7:20 am in Economics | Permalink

Here is a graph of planned infrastructure spending per person (by state) against the state unemployment rate.  The spending doesn't look especially targeted.  But would more targeting be better?  Not if that meant less useful spending.  The unemployment rate in Michigan is very high but that doesn't mean it's a good idea to spend a lot of money building schools in Detroit.

Fiscal stimulus would work better if the useful spending potential happened to be in the places where the unemployment rate was high but there's no reason to expect that to be the case.  Every recession requires reallocation.  

InfrastructureUnem
The data is from ProPublica, you can find more there although for technical reasons (see the extension if you are interested) I graphed infrastructure spending per person rather than per unemployed worker.

The ProPublica site maps/graphs infrastructure spending per unemployed worker against the unemployment rate but in effect that puts the number of unemployed people on both sides of the regression/graph and if there is any measurement error this can result in bias.  The graph with spending per unemployed worker is similar to the above but with a slightly more negative slope.

1 MostlyAPragmatist February 18, 2009 at 9:16 am

I don’t understand your point about targeting and re-allocation. When I look at this, I see that disproportionately large per capita infrastructure spending is going to states with small populations. That’s all I see and I don’t find it surprising.

2 Robert February 18, 2009 at 9:26 am

Hm. I recall reading that DC and the surrounding area had one of the lowest unemployment rates in the country. Guess I was wrong.

3 Chris February 18, 2009 at 9:45 am

Wow, this really looks like an “Empty and Northern” effect. Looks like the results are being driven by four contiguous unpopulated outliers. AK and VT are also unpopulated and northern as well…and DC…we could have guessed that one. I’m worried that the results are being driven by these few outliers.

4 Sarah February 18, 2009 at 10:12 am

Senate malapportionment guarantees that when the Senate writes a formula to distribute federal dollars to the states (such as infrastructure spending in the stimulus bill), each state’s allotment will not be proportional to its population. Low population states always secure the highest per capita dollars when the Senate writes a formula. You will also find that state Medicaid funding in the stimulus bill bears little relationship to state unemployment levels.

This isn’t a function of the inabilities of the Michigan delegation; it’s a function of the Constitution that grants equal voting power to states regardless of population.

Why didn’t the House insist on a formula that targeted dollars according to need? Think Ben Nelson (Nebraska), Susan Collins and Olympia Snowe (Maine)!

5 rrgg February 18, 2009 at 10:26 am

I haven’t read any discussion of this, but I always wondered if the long writer’s strike was going to exacerbate California’s budget problems. Has anyone estimated tax revenue losses for this? Maybe I’m overstating the case…

6 Vincent February 18, 2009 at 10:44 am

Maybe this will help clear the housing market as people move to where there are jobs. I think this because I think the fastest cure to the housing problem is the take to loss now.

7 Michael Martin February 18, 2009 at 11:36 am

I would say that California should consider secession, but with the direct democracy going here we wouldn’t last long.

8 John Thacker February 18, 2009 at 12:12 pm

Hm. I recall reading that DC and the surrounding area had one of the lowest unemployment rates in the country. Guess I was wrong.

The surrounding area is low, but DC itself is not. The DC Metro area, which includes a lot of non-DC, is low.

Like St. Louis proper, DC is a very small part of its own metro area.

9 John Thacker February 18, 2009 at 2:32 pm

Also, regarding DC, the per-capita spending is high, but that is likely because our infrastructure is used heavily by commuters from VA and MDCongressmen and their staff.

Fixed.

Of course, if DC only charged reasonable usage fees and congestion charges for the Metro and roads, it wouldn’t be such a problem.

10 anon February 18, 2009 at 3:35 pm

control for tax base?

11 Aris March 5, 2009 at 10:28 am

still try to connecting that to be politically correct, gvt consider to spending more initiative in infrastructure investment rather than reducing tax incentive .. mm

12 ZigoJaloflife March 24, 2009 at 1:36 pm

gwxb <a href=http://moodle.ttcps.tpc.edu.tw/user/view.php?id=9848&course=1>achat cialis</a> my site http://moodle.ttcps.tpc.edu.tw/user/view.php?id=9848&course=1 acheter cialis en ligne <a href=http://moodle.ttcps.tpc.edu.tw/user/view.php?id=9848&course=1>achat cialis</a> gp673

13 sex shop July 29, 2010 at 3:25 am

Wow. As a fairly libertarian Michigan resident, I didn’t think I could dislike the stimulus any more than I already did. But this graph does it. Thanks, Alex

Comments on this entry are closed.

Previous post:

Next post: