In This Recession, Teach Your Neighbors Well

Edward L. Glaeser is an economics professor at Harvard.

10:20 a.m. | Updated

This recession has hit some places much harder than others, and the spillovers from well-educated neighbors can help explain why.

The unemployment rate is 18.9 percent in Modesto, Calif., and 15.6 percent in Detroit, but only 7.7 percent in Minneapolis. Human capital "” as measured by the share of adults who have college degrees "” is among the best predictors of metropolitan fortunes, and, as the figure shows, that remains true today.

But the link between education and January 2010 unemployment is so strong that it creates something of a puzzle.

At first blush, it seems entirely natural that the share of adults with bachelor's degrees can explain about one-fifth in the variation in metropolitan area unemployment rates. After all, in January 2010, the unemployment rate was 17.6 percent for high school dropouts and 5.1 percent for college graduates. (For comparability with the metropolitan area data, which are not seasonally adjusted, these numbers are also not seasonally adjusted.) But the metropolitan unemployment rate declines too steeply by education to be explained simply by the fact that unemployment rates are lower for college graduates.

If the unemployment rate of an area just reflected the natural tendency of more educated people to have less unemployment, then local unemployment should follow the following formula:

Local Unemployment Rate=(Percent with B.A.'s) x (National Unemployment Rate for College Graduates) + (Percent Who Are High School Dropouts) x (National High School Dropout Unemployment Rate) +(Percent with High School Degrees but not College Degrees) x (National Unemployment Rate for that Group)

That formula is the unemployment rate predicted by the educational composition of an area and the national unemployment rate of different education groups.

Using unemployment rates from January 2010, and education shares from the last census (the best I could get), I multiplied the share of each area's population with college degrees by 5.1 percent; the share of each area's population who are high school dropouts by 17.6 percent; and the remaining share of each area's population by 10.25 percent.

The next figure shows the relationship between this predicted unemployment rate and the actual unemployment rate.

If the formula correctly predicted local unemployment, then local unemployment rates should be on the line, which is the 45 degree ray. If each area had exactly the unemployment level predicted by its education levels and national unemployment rate by education, then actual unemployment should rise one-for-one with predicted unemployment.

But areas with low levels of education, and hence high levels of predicted unemployment, are generally far above the line. Areas with high levels of education, and hence low levels of predicted unemployment, are generally below the line. As the predicted unemployment rate increases by one percent, the actual unemployment rate goes up by 1.8 percent.

This is a puzzle.

The more than one-for-one relationship between metropolitan area unemployment and the rate predicted by educational composition is an example of what economists call “social multipliers,” which may exist when aggregate relationships are stronger than individual relationships. In this case, the aggregate or metropolitan area relationship between unemployment and education is stronger than the individual relationship between unemployment and education.

Social multipliers may occur when one person's actions, like being unemployed or getting educated, influence everyone else.

If one layoff reduces the demand for another person's work and that leads to another layoff, then the impact of anything that increases unemployment will be multiplied. This is a standard Keynesian argument, but there are reasons to suspect that this isn't the story behind the overly strong relationship between local unemployment and local education.

The demand for most workers' output shouldn't be shaped by extremely local conditions: Demand is often driven by the state of the country or the world. Moreover, other variables that also predict unemployment at the individual level don't seem to generate the same multiplier that we see for education.

A stronger explanation for the social multiplier that exists between education and unemployment is the power of "Human Capital Spillovers," which is econ-speak for the benefits of having well-educated neighbors.

In 1993, James Rauch wrote a seminal paper showing that holding individual education constant, wages rise with the skills of metropolitan areas. Enrico Moretti has taken over this topic and written sophisticated papers that look both across metropolitan areas and within firms, showing that supermarket workers get more productive when better workers are in their shift.

Human capital spillovers can explain the overly strong metropolitan area relationship between skills and unemployment. If skilled co-workers make your company more productive, then your company will have fewer layoffs. If skilled entrepreneurs find opportunity in stormy times, then their hiring may keep unemployment rates down.

The fact that education has mattered so much during this recession only reminds us that America's future depends on its human capital.

Update: An earlier version of this post had an error in the metropolitan area predicted unemployment rate formula; “college graduation rates” should have said “unemployment rate.” Thanks to readers who pointed this out.

This recession has hit some places much harder than others, and the spillovers from having educated neighbors can help explain why, an economist writes.

The answer is complicated, at least according to the academic research on the subject.

Parents aren't just raising adorable kids. They are also producing little human capital units that are likely to grow up, get jobs, pay taxes and raise little human capital units of their own, an economist writes.

How will economic pressures influence Americans’ thinking on social issues like drug use, gambling, same-sex marriage and immigration?

China’s ascension to top economic dog is actually a re-ascension.

Canada’s concentrated megabanking system is often held up as a model for why Americans shouldn’t worry about their own large banks. Two economists explain why this thinking is wrong.

An economist recommends some good resources to help people understand the health care reform legislation signed into law.

Means-tested government benefits, and the "resentment zone" that goes with them, are more prevalent than ever, thanks to the housing crash. Programs like these delay the housing market's recovery, an economist writes.

For those unsatisfied with government efforts to scale back bankers’ compensation, there’s always “Whack-a-Banker,” a new arcade game.

Ireland is a cautionary tale, not a model, for Greece and other countries facing major debt issues, two economists write.

Catherine Rampell is the economics editor at nytimes.com.

David Leonhardt writes the Economic Scene column, which appears in The Times on Wednesdays.

Sewell Chan writes about economic issues from Washington D.C.

Marc Lacey is The Times's bureau chief for Mexico, Central America and the Caribbean.

Economists offer readers insights about the dismal science.

Economics doesn’t have to be complicated. It is the study of our lives "” our jobs, our homes, our families and the little decisions we face every day. Here at Economix, David Leonhardt, Catherine Rampell and other contributors will analyze the news and use economics as a framework for thinking about the world. We welcome feedback, at economix@nytimes.com.

In order to view this feature, you must download the latest version of flash player here. var so = new SWFObject('http://graphics8.nytimes.com/packages/flash/business/20081008-credit-charts-graphic/credit3.1.swf','nytSWF',337,385,9,'#FFFFFF'); so.addParam('allowScriptAccess','always'); so.addParam('allowFullScreen','true'); so.addParam('BASE','http://graphics8.nytimes.com/packages/flash/business/20081008-credit-charts-graphic/'); so.addVariable('allowCaching',true); so.write('embed70'); #embed70{visibility:visible !important;}h2.multiHeadline {font-size:156% !important;padding:0px 0px 7px !important;margin:6px 6px 11px 3px !important;border-bottom:1px solid #CCCCCC !important;}

Introducing

Apture allows readers to dig deeper into a subject without ever leaving the blog post. When you click on any link marked by the icons , , or , you will be able to view video, reference materials, images and other related media. Please e-mail your feedback and thoughts on this feature to apture@nyt.com.

Multimedia Breaking Down the Bailout

An accounting of the government’s rescue package.

How the Government Dealt With Past Recessions

Three economists explain what worked and what didn't.

Geography of a Recession

A map of unemployment rates across the United States, now through January.

Related Article In order to view this feature, you must download the latest version of flash player here. var so = new SWFObject('http://www.nytimes.com/packages/flash/multimedia/swfs/AS3Multiloader.swf','nytSWF',165,104,9,'#FFFFFF'); so.addParam('allowScriptAccess','always'); so.addParam('allowFullScreen','true'); so.addParam('wmode','opaque'); so.addParam('BASE','http://graphics8.nytimes.com/packages/flash/us/20090300-economy-user-photos/promos/'); so.addVariable('contentPath','http://graphics8.nytimes.com/packages/flash/us/20090300-economy-user-photos/promos/163quilty.swf'); so.addVariable('allowCaching',true); so.write('embed443'); #embed443{visibility:visible !important;}

Faces, numbers and stories from behind the downturn.

Special Features The Debt Trap

A series about the surge in consumer debt and the lenders who made it possible.

The Reckoning

A series exploring the origins of the financial crisis, from Washington to Wall Street.

Blogroll Blogroll Barking Up the Wrong Tree Brad DeLong Cafe Hayek Calculated Risk Capital Gains and Games Dani Rodrik's Weblog DataPoints: The Dismal Scientist Blog Econbrowser EconLog Economist's View Greg Mankiw Marginal Revolution Nouriel Roubini's Global EconoMonitor Nudge Robert Reich Tax.com TaxProf TaxVox The Baseline Scenario The Becker-Posner Blog The Big Picture Will Wilkinson: The Fly Bottle Blogs From Newspapers and Magazines Andrew Sullivan Ezra Klein Felix Salmon Floyd Norris Freakonomics Free exchange (The Economist) James Surowiecki Megan McArdle Money-Supply (Financial Times) Paul Krugman Real Time Economics (WSJ) Wolf Forum (Financial Times) Economic Resources Employment Statistics GeoFRED: Geographic Federal Reserve Data Historical Data on Job Growth and Wages Historical Unemployment Data Inflation Calculator Interactive Housing Calculator International G.D.P. Rankings Latest Job Market Data Local Gas Prices Statistics on Income, Poverty and Health Insurance Coverage in the U.S. U.S. G.D.P. Statistics Tag List DAILY ECONOMIST 344 HEALTH CARE 189 UNEMPLOYMENT 160 TAXES 146 HEALTH INSURANCE 98 EMPLOYMENT 94 UWE E. REINHARDT 88 EDWARD L. GLAESER 85 STIMULUS 82 HOUSING 80 CASEY B. MULLIGAN 71 JOBS REPORT 71 RECESSION 67 U.S. HEALTH CARE COSTS 65 BAILOUT 61 CREDIT CRISIS 59 JOBS 58 NANCY FOLBRE 57 SIMON JOHNSON 57 WOMEN IN THE WORKFORCE 56 FEDERAL RESERVE 56 BANKS 53 INTERNATIONAL ECONOMICS 50 FINANCIAL CRISIS 48 STOCKS 47 OPEN MARKET 46 BUDGET DEFICIT 46 GROSS DOMESTIC PRODUCT 45 EDUCATION 45 HIGHER EDUCATION 44 Archive Select Month March 2010 February 2010 January 2010 December 2009 November 2009 October 2009 September 2009 August 2009 July 2009 June 2009 May 2009 April 2009 March 2009 February 2009 January 2009 December 2008 November 2008 October 2008 September 2008 Home World U.S. N.Y. / Region Business Technology Science Health Sports Opinion Arts Style Travel Jobs Real Estate Autos Back to Top Copyright 2010 The New York Times Company Privacy Terms of Service Search Corrections RSS First Look Help Contact Us Work for Us Advertise Site Map function apture_onload() { apture.addCallback(apture.WINDOW_OPEN, function() { dcsMultiTrack('DCS.dcssip','www.nytimes.com','DCS.dcsuri','/blogs/Apture.html','WT.ti','Apture','WT.z_dcsm','1'); }); } if (typeof NYTD.Blogs.user != 'undefined') { if(NYTD.Blogs.user.isLoggedIn()) { var dcsvid=NYTD.Blogs.user.getId(); var regstatus="registered"; } else { var dcsvid=""; var regstatus="non-registered"; } } var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www."); document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E")); var pageTracker = _gat._getTracker("UA-4406282-67"); pageTracker._initData(); pageTracker._trackPageview(); if((document.referrer).indexOf(document.location.hostname) == -1) { var referrer = document.referrer; } else { var referrer = ''; } document.write('');

Apture allows readers to dig deeper into a subject without ever leaving the blog post. When you click on any link marked by the icons , , or , you will be able to view video, reference materials, images and other related media. Please e-mail your feedback and thoughts on this feature to apture@nyt.com.

An accounting of the government’s rescue package.

Three economists explain what worked and what didn't.

A map of unemployment rates across the United States, now through January.

Faces, numbers and stories from behind the downturn.

A series about the surge in consumer debt and the lenders who made it possible.

A series exploring the origins of the financial crisis, from Washington to Wall Street.

Read Full Article »