Paths to Prosperity - Knowledge is Key for Fourth District States

Sep 21, 2026

Since 1939, real U.S. per capita personal income has risen sharply, yet income levels still differ across states. The document explains how persistent gaps narrowed through “convergence” but remain significant by 2004. It attributes remaining differences to identifiable factors linked with state knowledge stocks—particularly patent innovativeness and educational attainment—along with industry structure. It also outlines the estimation approach and a five-year lag strategy to ensure growth factors influence income rather than reverse causality.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages5

Reading Time13 min

AccessFree to download

Paths to Prosperity: Knowledge is Key for Fourth District States

How Are the Effects of Growth Factors Estimated?

by Paul W.Bauer and Mark E.Schweitzer Since 1939,real U.S.per capita per-sonal income has grown more than 400percent—an average of 2.5 percent peryear—greatly raising living standards.Each state has seen its per capita incomerise dramatically (see figure 1),but notall states have grown at the same rate.Between 1939 and 2004,these differinggrowth rates have caused a substantialreduction in the income differencesbetween high-and low-income states,aconcept known as convergence. For example,the highest-income state in1939(Delaware)had a per capita per-sonal income that was four and a halftimes higher than that of the lowest-income state(Mississippi).Since themid-1970s,per capita personal incomein the highest-income states has typi-cally been only slightly less than twotimes the income level of the lowest-income states.While these differencesare clearly smaller than those in the1930s,current income differencesremain significant:Connecticut's 2004top-ranked per capita personal income(S45,566)was almost double that ofMississippi's(S24,397)(see figure 2). What are the sources of these remainingdifferences?Recent research by Bauer,Schweitzer,and Shane find that there areidentifiable factors reliably associatedwith income differences in U.S.statesThey find that a state's knowledgestocks—the innovativeness of its firmsand the education and training of itsworkforce—are key determinants ofits per capita personal incomeIn thisEconomic Commentary,we focus on theperformance of Fourth District states inthese areas.The results highlight issuesthat policymakers must face if they wantto raise the income levels of FourthDistrict states. # Paths to Prosperity:Knowledge is Keyfor Fourth District States ■ How Are the Effects ofGrowth Factors Estimated?Bauer,Schweitzer,and Shane apply astatistical model to identify factors thatlead to persistent differences in state percapita personal incomes.The key fea-tures that help produce reliable estimatesare as follows: ●Information on income levels overmany years—65 years of state percapita income for the lower 48 states ·Information on how a wide variety ofpotential growth factors—educationalattainment,patents,industry struc-ture,taxes,infrastructure spending,banking deposits,and climate—differbetween states ·A statistical model consistent withgrowth theory—the estimated modelaccounts for the income changes fromthe unobserved flows of capital andlabor(see Barro and Sala-i-Martin,and Islam) ·A strategy ensuring that the factorsare influencing income and not viceversa—the data on factors is fiveyears prior to the income data This last point requires more elabora-tion.When estimating the effects ofthese possible growth factors,it isnecessary to carefully control for thepossibility that per capita personalincome might in turn influence thefactors.For example,states with highercurrent incomes are likely to choose tospend more on education.The authorsstatistically verify that a five-year lag isenough to ensure that the factors'effectson income are captured,not currentincome's effect on the factors. August 15,2006 Even as per capita income hasincreased across the United States,differences among states'incomesremain.What are the sources ofthese remaining differences?ThisCommentary identifies and analyzesthe key factors—patents,educationalattainment,and industry structure—that influence income-growth ratesand thus per capita incomes.It alsoexplores where the Fourth Districtfalls in relation to other states andthe country as a whole. The key identified factors turn out to bethe knowledge stocks of patents andeducational attainment,as well as astate's industry structure.Knowledgestocksare measured by three variables:The first measures the innovativeness ofa state's firms.Data on patents issued toa state are used because they are highlycorrelated with research and develop-ment spending,but unlike these moredirect measures,they are available at thestate level further back in time(seeGriliches).Th

FAQs

What explains the remaining income differences among U.S. states by the early 2000s?

The document points to knowledge stocks tied to patent innovativeness and educational attainment, as well as the state’s industry structure.

How do the authors estimate the effects of growth factors on income?

They use a statistical model consistent with growth theory and rely on long-run income data plus cross-state measures of multiple potential growth factors.

Why does the document emphasize a five-year lag in the data?

The lag helps control for reverse causality, so the estimated effects reflect how factors influence income rather than how current income changes those factors.

Sep 21, 2026

Uploaded by Clementine