Human capital, social capital, and public schooling

*Corresponding author. Tel.: 972-7-6472288; fax: 972-7-6472941.E-mail addresses: [email protected] (M. Gradstein), [email protected]

(M. Justman)

European Economic Review 44 (2000) 879}890

The Political Economy of Education

Human capital, social capital, andpublic schooling

Mark Gradstein*, Moshe Justman

Department of Economics, Ben Gurion University, P.O. Box 653, 84105 Beer Sheva, Israel

Abstract

Public education contributes to growth not only by building human capital but also byinstilling common norms that increase social cohesion. This is modeled in the context ofa political economy framework in which social cohesion reduces wasteful rent seeking,and thus strengthens incentives for investment in human captial. The political decisionsthat determine whether di!erent social groups retain separate schooling systems, oradopt an integrated system, weigh these material advantages against the psychic cost toparents of alienating their children from traditional values. This aspect of public educa-tion helps explain why, commonly, education is publicly administered as well as publicly"nanced. ( 2000 Elsevier Science B.V. All rights reserved.

JEL classixcation: I21; H42; O15; D72

Keywords: Public education; Human capital; Social capital

1. Introduction

Although education does not have the technical attributes of a public good} it is both appropriable and divisible } public education, especially at theprimary and high school levels (K-12), enjoys wide political support in almost allcountries. Several recent e!orts seek to explain this in terms of the instrumental

0014-2921/00/$ - see front matter ( 2000 Elsevier Science B.V. All rights reserved.PII: S 0 0 1 4 - 2 9 2 1 ( 9 9 ) 0 0 0 4 4 - 6

1This applies both to the proponents of endogenous growth theory, from Romer (1986) and Lucas(1988) on, and to those who question its necessity, e.g., Mankiw et al. (1992).

2Papers in this vein which focus on the political economy of education include Glomm andRavikumar (1992), Boldrin (1993), Saint-Paul and Verdier (1993), Benabou (1996), Gradstein andJustman (1997), and Fernandez and Rogerson (1999).

3Cf. Grossman and Kim (1997), who argue that an increase in a person's human capital makespredation less attractive. Therefore, well-endowed people can increase their own consumption bydiverting some of their income to educating the less endowed.

4Of course, heterogeneity also has its advantages: a diversity of perspectives can be mutuallyenriching. Nevertheless, there is evidence that in some contexts the overall e!ect is detrimental,possibly even destructive. Recent events in former Yugoslavia are a tragic case in point.

role of education in building human capital, now widely recognized as anessential production factor of national output.1 Studies in this vein highlight thepotential bene"ts of government intervention as a means of internalizing theexternal bene"ts of education, relaxing credit constraints, and redistributingincome.2 Yet these are essentially arguments for public xnancing of education,rather than for public provision. The bene"ts they attribute to public educationcan be realized through less intrusive means than public administration, e.g.,through the use of subsidies or vouchers. This suggests that there may be otheradvantages to public provision of education, largely neglected in this literature,which might better serve to explain its broad support. These advantages, weargue in this paper, do not derive from the instrumental role of education inbuilding human capital, through the transmission of knowledge and skills, butfrom its normative role in building social capital. Public schooling instillscommon cultural norms and ethical values that lower economic transactioncosts and reduce social tensions between di!erent population groups. Thisnormative aspect of education requires the direct controls of a publicly adminis-tered schooling system (Lott, 1990; Kremer and Sarychev, 1998).

The economic bene"ts of normative education work in a variety of ways.Instilling civic virtues from an early age through public schooling reduces futureenforcement costs.3 Relatedly, uniform public schooling in a common culturegenerates network externalities by reducing transaction costs and thus facilitat-ing economic activity } di!erences in language, custom or religion can give riseto misunderstandings that undermine the e$ciency of production and exchange.(We examine this aspect in Gradstein and Justman (2000).) Yet a third bene"t ofsocial cohesion derives from the potential for redistributional con#ict amongdistinct ethnic, cultural or religious groups. Uniform public schooling is a meansby which the parent generation can e!ectively reduce the likelihood of suchcon#ict in the following generation. Collectively, parents can contribute to theeconomic welfare of their children by helping them assimilate within a broadercultural framework. But this is not without cost: the speci"c social capital of theparents is sacri"ced, and the traditional values in which they were raised arediluted, weakening the link between parent and child.4

880 M. Gradstein, M. Justman / European Economic Review 44 (2000) 879}890

5There are detailed injunctions regarding class size, restrictions on the harshness of punishment,and a vivid discussion of whether a teacher who covers a lot of ground but makes mistakes ispreferable to one who is slower but more precise.

6 It may have contributed to commercial ties among medieval Jewish traders, recently studied byGreif (1989).

This emphasis on the normative role of education is strongly re#ected inhistorical experience. The role of education in building social capital was oftena prominent motive for developing public education systems. The oldest re-corded system of universal (male) public education, dating back nearly twothousand years, is ascribed in the Talmud to the High Priest Joshua b. Gamla, inthe generation preceding the destruction of the Second temple, &2 for but forhim the Torah would have been forgotten from Israel. For at "rst if a child hada father his father taught him, and if he had no father he did not learn at all

2 At length Joshua b. Gamla came and ordained that teachers of youngchildren should be appointed in each district and each town, and that childrenshould enter school at the age of six or seven'. (Babylonian Talmud, BabaBathra, 21a). The children were taught to read, understand and memorizeBiblical texts, mostly from the Torah. Teachers were under the supervision ofthe rabbinical court and the scholars, and could be dismissed without warning ifthey failed to perform their duties adequately.5 Parents were expected to pay fortheir children's instruction, but needy cases were supported from the publicpurse (Steinsaltz, 1989); as with subsequent systems of religious education,public administration of the education system, rather than public "nance, wasthe predominant mode of intervention. This education system, which persistedin varying forms throughout the Jewish Diaspora, clearly had an instrumentaldimension in its contribution to literacy levels, but this was an ancillary e!ect.Its primary contribution was to social capital, creating a community of normsand values, and a system of adjudication.6

The earliest examples of modern public education date back to the eighteenthcentury, established under the absolute monarchies of Frederick the Great inPrussia and Maria-Theresa in Austro-Hungary (Lamberti, 1989). Building onan existing infrastructure of voluntary religious education, their purpose was toinstill duty and devotion in their subjects. Other regimes, subscribing to di!erentagendas, similarly sought to instill a sense of civic duty and internalize ethicalnorms, and thus reduce the cost of enforcing the rule of law. In 1833 the LoiGuizot laid the foundation for a centralized, highly regimented education systemin France. Jardin (1983, p. 113) describes its purpose: &The school reform,conceived at a time when France was preparing to make a new beginning and totrain future generations in a new manner, was designed to foster social cohesion

2'. In both Prussia and France public supervision and control of schoolingpredated full public "nancing by several decades (Green, 1990). In both, the widedissemination of a secular ideology is credited with playing a key role inindustrialization (Gerschenkron, 1962).

M. Gradstein, M. Justman / European Economic Review 44 (2000) 879}890 881

In democracies, literacy is essential in itself for the informed behavior of theelectorate, on which the proper functioning of democracy relies. This linkbetween democratization and education is especially apparent in England,where public schooling followed closely on the extension of the politicalfranchise to the working classes in the Second Reform Act of 1867 and the ThirdReform Act of 1884}88. With the Education Act of 1870, &the education of theEnglish poor became for the "rst time the direct concern of the nation' (Dicey,1914, p. 277). Here, too, public schooling was made compulsory more than tenyears before the legislation of public responsibility for its "nancing.

This role of education as an integrating force shaping the modern industrialnation-state was accentuated in countries that attracted large numbers ofimmigrants (Green, 1990). Thus the spread of public primary education in theUnited States in the mid-nineteenth century was aimed at helping newly arrivedimmigrants acquire not only language skills but also new social and politicalnorms, and the discipline of the workplace in the modern industrial economy(Edwards and Richey, 1963; Bowles and Gintis, 1976). Similarly, in modernIsrael the integration of an immigrant society was a central goal of the publicschool system.

Public education has also played an important role in the process of nationbuilding in many multi-ethnic developing countries that gained their indepen-dence after World War II. Singapore is a case in point. Under colonial rule,education in Singapore was ethnically segregated, its level and content varyingacross population groups, but after independence the government uni"ed thedi!erent education streams and aggressively promoted universal public school-ing, with English as a common o$cial language. This facilitated communicationbetween the di!erent ethnic groups, and while cultural di!erences remain,potentially destructive ethnic con#ict has been avoided (Thomas et al., 1980). Incontrast, some of the more ethnically diverse African countries were unable to"nd a common ground on which to base a uni"ed school curriculum } in termsof language, culture, and social norms. In these countries, where public educa-tion was not successful in promoting social cohesion, a high degree of mistrustamong di!erent ethnic groups remains a misfortunate reality, often #aring up inviolent con#ict (Easterly and Levine, 1997).

The statistical signi"cance of the contribution of social capital to growth wasrecently demonstrated by Knack and Keefer (1997), La Porta et al. (1997), andTemple and Johnson (1998), reviving an approach "rst set out by Adelman andMorris (1967). These studies show that high levels of trust and social participa-tion are positively correlated with growth, after controlling for other growth-promoting factors. Conversely, ethnic heterogeneity in populations comprisingrival groups harboring a large degree of mutual mistrust and animosity is oftenassociated with costly struggles over power and resources that are detrimentalfor growth. In extreme cases these take a violent form, as in the recent tragicexperience in former Yugoslavia. But the adverse e!ects of heterogeneity can be


7Tornell and Lane (1999) highlight the detrimental e!ect on growth of con#icts deriving fromcompetitive rent seeking among ethnic groups, but do not consider the potential role of publiceducation in smoothing out these tensions.

8 In reality, of course, home environment is also an important input in a child's education, andalthough public schooling emphasizes common norms, the result is never entirely homogeneous.Thus a child might learn a local dialect at home in addition to the o$cial language taught at school.

experienced without explicit military con#ict, e.g., through nonviolent politicalstruggle over resources between rival ethnic groups. Thus Alesina et al. (1997)argue that ethnic heterogeneity leads to a lower supply of public goods in localcommunities in the US; and Alesina and La Ferrara (1999) "nd that socialparticipation is lower in ethnically fragmented communities.

The present paper integrates these various strands of the literature by theoret-ically elaborating the contribution of public schooling to growth through its rolein reducing redistributive rent seeking between competing ethnic groups.7 Therent-seeking activities that arise when separate school systems accentuate ethnicdi!erences reduce the anticipated returns to schooling and thus dampen invest-ment in human capital. Social cohesion thus doubly promotes growth: byreducing wasteful rent-seeking activities and increasing investment in produc-tive human capital. The political decisions that determine whether a singleuniform education system is adopted, or whether each ethnic group has its ownschool system, weigh these economic advantages against the psychic cost ofalienating one's child from own's one traditions and values.

2. Assumptions

We assume a two-period economy populated by a continuum of householdseach consisting of a parent and a child. Parents are exogenously endowed withsocial characteristics } language, culture, ideology. Assume for simplicity thatthe parent generation is divided into just two uniform groups, &reds' and &greens',denoted by the index j"r, g, and that the reds constitute a majority. Denote theproportion of reds to greens by c(c'1), so that the measure of individuals ineach period is 1#c. Assume that social characteristics are captured by a singleparameter, d

j, j"r, g, where 04d

j41, and that the two groups have di!erent

characteristics, drOd

g. (We will use culture and ideology interchangeably when

talking about these social characteristics.) To focus on essentials assume furtherthat all parents are endowed with the same initial income, w.

Parents decide on the cultural orientation of their children's schooling, and onthe division of their initial endowment between current consumption andinstrumental investment in their children's human capital. Thus schoolingendows the child with a level of human capital, h

j, and a social orientation,

which we denote sj, j"r, g.8 The cultural distance between parent and child is


9See Bisin and Verdier (1997) for a more detailed economic perspective on the disutility of havinga child di!erent from oneself.

10Equivalently, parents whose choose to &opt out' of public schooling are assumed to receive a fulltax credit. One might also consider alternative regimes in which parents receive a partial tax creditfor opting out of public education or no credit at all, as is the case in the US. For our presentpurpose, the "nancing of public education can be viewed as a constitutional decision, determined atan earlier stage of the political process by the relative power of the di!erent social groups. This issueis not discussed here, but deserves further study.

11There are other assumptions one could make regarding the type of alternative system that isadopted, e.g., one could have community school systems for either or both groups.

12An alternative source of ine$ciency associated with heterogeneity, which we do not pursuehere, derives from the adverse impact on productivity of communication di$culties between peopleof diverse cultural backgrounds.

13For notational simplicity, the technology parameter is set uniformly equal to 1, but in general, itmay di!er across groups or individuals.

then Ddj!s

jD, and we assume that reducing this distance is directly bene"cial to

parents' utility.9 To "x ideas, assume parents' utility has the following quasi-linear form:

;(cj1

, cj2

,Ddj!s

jD)"u(c

j1)#c

j2!Dd

j!s

jD (1)

where cj1

denotes current consumption, cj2

denotes next-generation consump-tion, and u is monotonically increasing and concave.

We restrict our attention to two possible school systems: a private system thatpreserves parents' distinct cultural identities in the next generation, and a singleuniform school system that erases them. We assume that the majority redsdetermine the social orientation of the public education system and its level ofinvestment in human capital, but cannot coerce the minority greens to partici-pate in it or pay for it.10 In other words, the reds o!er a social orientation, d, andhuman capital level h, and the greens can either accept or reject it. If the greensaccept the reds' o!er, we have uniform, universal public schooling on these terms;if they reject it, we have private schooling.11 While private schooling o!ers theadvantage of minimizing the cultural distance between parent and child, it doesso by perpetuating distinct cultural identities which promote ine$cient redis-tributive rent-seeking activities between the two group in the next generation,each group marshalling wasteful e!ort to gain what it can. A single uniformschool system allows these e$ciency losses to be avoided.12

Whichever school system is chosen, parents fully "nance the schooling of theirown children from their initial endowment, allocating it between current con-sumption, c

j1, and investment in human capital, h

j:

w"cj1#h

j. (2)

This investment determines next-period gross income, yj, through a linear

production technology,13

yj"h

j. (3)


However, net income, which equals next-generation consumption, is also a!ec-ted by redistributive rent-seeking activities determined by the social capital thatparents pass on to their children.

In explicitly modeling these redistributive rent-seeking activities we followa standard approach (see, e.g., Becker, 1983). We make the simplifying assump-tion that each of the two groups is able to completely overcome internal freeriding incentives, so that each collectively determines the amount of rent-seekinge!ort it wishes to expend. We assume that there is a share, 0(a(1, ofsecond-period individual gross income, which is protected by the prevailingsystem of property rights, with the remaining unprotected share exposed to rentseeking. Recalling that gross income is equivalent to human capital, ah

jrepres-

ents the amount of protected income of an individual of group j; and (1!a)H isthe total amount of second period income available for redistribution, whereH"ch

r#h

gdenotes the total amount of investment in human capital.

Let xrand x

gdenote the respective rent seeking e!orts of individuals in each

of the two groups. Let S(cxr, x

g) denote the proportion of the unprotected share

of national income that accrues to the red group, where S is increasing in its "rstargument and decreasing in its second; then 1!S(cx

r, x

g) is the proportion that

accrues to the green group. To simplify the derivations we assume that S takesthe functional form S(cx

r,x

g)"cx

r/(cx

r#x

g), and posit S(0, 0)"c/(c#1).

Thus when all members of the two groups exert identical positive e!ort the morenumerous reds are able to capture a larger fraction of resources. The proceedsfrom redistribution are allocated equally between group members. Finally, letbjdenote the marginal cost which group j incurs in mobilizing resources for rent

seeking purposes. We assume that it is inversely related to group size, withbr"1/c, and b

g"1. With these assumptions, and recalling Eq. (3), net income,

hence consumption, in the second generation is given by

cr2"ah

r#(1!a)Hx

r/(cx

r#x

g)!x

r/c, (4a)

cg2"ah

g#(1!a)Hx

g/(cx

r#x

g)!x

g. (4b)

3. Analysis

3.1. Private schooling

We proceed backwards, starting with the analysis of redistribution in the nextgeneration. In the private schooling case, heterogeneity persists in the nextgeneration, so that each of the two groups exerts pressure to redistribute in itsfavor. They then choose x

rand x

gso as to maximize (4a) and (4b) respectively.

Straightforward analysis then reveals that, the di!erence in group size notwith-standing, each individual in the two groups exerts the same amount of rent-seeking e!ort: x

r"x

g"(1!a)Hc/(1#c)2. Substitution back into (4a) and (4b)


then yields

cr2"ah

r#(1!a)Hc/(1#c)2, (5a)

cg2"ah

g#(1!a)H/(1#c)2. (5b)

Anticipating the next-period struggle, parents invest in the human capital oftheir children, and determine their social orientation, so as to maximize theirown utility. Clearly, in this simple framework, there is no incentive under privateeducation to bring up one's child other than in one's own tradition, so s

j"d

jthe level of human capital investment is determined by each parent so as tomaximize utility (1), subject to the budget constraint (3), and anticipated next-period consumption (5). The resulting "rst order conditions at the internalequilibrium are

!u(w!hj)#a"0 (6)

for j"r, g. Let h0

denote the individually optimal level of investment deter-mined by (6), and note that it is the same for both groups. Nonetheless, themajority reds derive more second-period consumption, and hence greater utility,from their investment because of their advantage in rent seeking. Total di!eren-tiation of (6) reveals that investment is an increasing function of a: betterprotection of property rights increases overall investment. Letting ;

j0denote

the utility level of a group j parent under private schooling, we then obtain

;r0"u(w!h

0)#[a#(1!a)c/(c#1)]h

0, (7a)

;g0"u(w!h

0)#[a#(1!a)/(c#1)]h

0. (7b)

Di!erentiating (7) with respect to a, and invoking (6), we obtain that both groupsbene"t from stronger property rights. Summarizing:

Proposition 1. In a private school system better protection of property rightsincreases parental investment in instrumental schooling in both the majority and theminority, and increases parents' utility in both groups.

3.2. Public schooling

Now consider universal, uniform public education. Its social orientation, sH,and level of instrumental human capital, hH, are chosen by the red majority andaccepted by the green minority. As all children share a common culture there isno basis for wasteful rent-seeking e!ort. Consequently (and as all parents areendowed with the same initial income, w), there is unanimity regarding thedesired level of instrumental human capital, derived by maximizingu(w!h)#h, and characterized by the "rst-order condition

!u(w!hH)#1"0. (8)


It follows from Proposition 1 that hH'h0

(viewing public schooling as a limit-ing case of private schooling with full protection of property rights, a"1). Ash0'c

j2for j"r, g:

Proposition 2. There is more investment and higher future income under publicschooling than under private schooling.

The two groups are diametrically opposed regarding the social orientation ofthe public school system. Assume without loss of generality that d

r(d

g. Then

the reds prefer as low a value as possible within the interval [dr, d

g], while the

greens prefer as high a value as possible, and cannot be coerced into joiningthe public system. It follows that the values of s acceptable to minority parentsare bounded from below by the greater utility from material goods thatpublic schooling o!ers, compared with private schooling, and the lowest accept-able value of s is given by

u(w!hH)#hH!(dg!s)"u(w!h

0)#[a#(1!a)/(c#1)]h

0. (9)

If the value of s that solves (9) is less than drthen public schooling will surely be

established, as the red majority must bene"t from a public system that adopts itsown cultural orientation. This may occur if property rights (represented by a)are weak, the reds' majority, c, is large, and the cultural distance between redsand greens, d

g!d

r, is small. Otherwise, let sH denote the solution of (9). Then

public schooling is established if sH also represents an improvement over privateschooling for the red majority, i.e., if it satis"es

u(w!hH)#hH!(sH!dr)5u(w!h

0)#[a#(1!a)c/(c#1)]h

0. (10)

Combining (9) and (10), we "nd that public schooling is feasible if

u(w!hH)#hH!u(w!h0)!(1/2)(1#a)h

05(1/2)(d

g!d

r). (11)

Inspection of (11) reveals again that a larger cultural distance between the twogroups impedes the adoption of public schooling. Goldin and Katz (1998)provide some empirical support for this result. Focusing on the evolution ofpublic (secondary) schooling in the US between 1910 and 1940, they "nda strongly positive relationship between public schooling and measures of socialcapital, and, in particular, proxies for social homogeneity. James (1993) "ndssimilar e!ects in a cross sectional analysis of countries: those with strongerreligious divisions devote a larger share of resources to private education.Cohen-Zadeh and Justman (2000) "nd a similar e!ect among local communitiesin the US, after controlling for income levels and income distribution.

The relative power of the groups determines the orientation of the uniformpublic system. Eq. (9) implies that sH is decreasing in c: the smaller the minorityof the greens the greater their loss from rent-seeking con#ict and hence the lowerthe value of s which they are willing to accept, i.e., the more they are willing to


14The large share of private schooling in the Netherlands, for example, is consistent with such ane!ect.

15 In a related context, Gradstein (1999) argues for a similar role of public schooling when thesource of potential social tensions is income inequality.

concede regarding the cultural orientation of the public school system. Equiva-lently, the larger the majority of the reds the closer will the social orientation ofthe public system conform to their own ideology. However, a large majority alsogains more from rent-seeking behavior. Thus the relative size of the majoritya!ects the decision to adopt public schooling in two opposing ways; in thecurrent formulation the two e!ects cancel out, and relative size does not enter inequation (11). Finally, since the derivative of the left-hand side of (11) withrespect to a is negative (after applying the "rst-order condition in Eq. (6)),stronger property rights imply a smaller loss from rent-seeking activities andhence less of a gain from a uniform public school system.14 More generally,public schooling serves in this context as a precommitment device against futurecon#ict. If such con#ict can be averted by other means, culturally speci"cschools that allow parents to educate their children in their own tradition, arepreferable.15 Summarizing:

Proposition 3. 1. Public schooling is more viable the less polarization there isbetween the two groups.

2. The ideology of the public school system will be closer to that of the majoritythe larger is its share in the population.

3. The stronger the system of property rights the smaller the advantage of publicschooling.

4. Conclusion

The economic growth literature, while underscoring the importance ofschooling for the accumulation of human capital, has largely disregarded thecontribution of public education to social capital. Yet historians of educationsystems emphasize this aspect of schooling as a primary objective of publiceducation. Public education, especially in early stages of nation building, playsa key role in promoting social cohesion and reducing ethnic tensions. Moreover,recent evidence has shown that the measurable dimensions of social capital} observed levels of trust, voluntarism and social participation } have a signi"-cant positive association with growth. The present paper o!ers a formal modelof this aspect of public education, describing a twofold e!ect of social cohesionon growth. A lessening of social tensions reduces wasteful rent-seeking activities,which increases the returns to human capital and promotes higher levels ofinvestment in education. In this context, political support for public education


re#ects a balance between the material advantages of a common social fabricand the psychic cost to parents of relinquishing their speci"c cultural heritage.Hopefully, a better understanding of the role of education in building socialcapital can shed new light on the historical origins of public education, and onits potential contribution to economic welfare.

References

Adelman, I., Morris, C.T., 1967. Society, Politics, and Economic Development. Johns HopkinsUniversity Press, Baltimore, MD.

Alesina, A., Baqir, R., Easterly W., 1997. Public goods and ethnic divisions. Quarterly Journal ofEconomics, forthcoming.

Alesina, A., La Ferrara, E., 1999. Participation in heterogeneous communities, Working Paper 7155,NBER, Cambridge, MA.

Babylonian Talmud. English translation. Soncino, London.Becker, G.S., 1983. A theory of competition among pressure groups for political in#uence. Quarterly

Journal of Economics 98, 371}400.Benabou, R., 1996. Heterogeneity, strati"cation, and growth: Macroeconomic implications of

community structure and public "nance. American Economic Review 89, 584}609.Bisin, A., Verdier, T., 1997. The economics of cultural transmission and the dynamics of preferences.

Working Paper 9703. DELTA, Paris.Boldrin, M., 1993. Public education and human capital accumulation. Discussion Paper 1017,

Kellogg Graduate School of Management. Northwestern University, Evanston, IL.Bowles, S., Gintis, H., 1976. Schooling in Capitalist America: Educational Reform and the Contra-

dictions of Economic Life. Basic Books, New York.Cohen-Zadeh, D., Justman, M., 2000. Demand for public and private education: Theory and

evidence, Monaster Center Discussion Paper, Ben Gurion Univeristy, Beer Sheva.Dicey, A.V., 1914. Lectures on the Relation Between Law and Public Opinion in England During

the Nineteenth Century, 2nd ed. Macmillan, London (reissued in 1962).Easterly, W., Levine, R., 1997. Africa's Growth Tragedy: Policies and Ethnic Divisions. Quarterly

Journal of Economics 112, 1203}1250.Edwards, N., Richey, H.C., 1963. The School in the American Social Order. Houghton Mi%in,

Boston, MA.Fernandez, R., Rogerson, R., 1999. Equity and resources: An analysis of education "nance systems.

Working Paper 7111. NBER, Cambridge, MA.Gerschenkron, A., 1962. Historical Backwardness in Historical Perspective. Harvard University

Press, Cambridge, MA.Glomm, G., Ravikumar, B., 1992. Public versus private investment in human capital: Endogenous

growth and income inequality. Journal of Political Economy 100, 818}834.Goldin, C., Katz, L.F., 1998. Human capital and social capital: The rise of secondary schooling in

America. 1910 to 1940, Working Paper 6439. NBER, Cambridge, MA.Gradstein, M., 1999. An economic rationale for public education: The value of commitment. Journal

of Monetary Economics, forthcoming.Gradstein, M., Justman, M., 1997. Democratic choice of an education system: Implications for

growth and income distribution. Journal of Economic Growth 3, 169}183.Gradstein, M., Justman, M., 2000. Education, social cohesion and economic growth. Monaster

Center Discussion Paper, Ben Gurion University, Beer Sheva.Green, A., 1990. Education and State Formation. St. Martin's Press, New York.


Greif, A., 1989. Reputation and coalitions in medieval trade: Evidence on the Maghribi traders.Journal of Economic History 49, 857}882.

Grossman, H., Kim, M., 1997. Human Capital and Predation: A Positive Theory of EducationPolicy. Working paper, Brown University, Providence, RI.

Lamberti, M., 1989. State, Society, and the Elementary School in Imperial Germany. OxfordUniversity Press, Oxford, London.

James, E., 1993. Why do di!erent countries choose a di!erent public-private mix of educationalservices? Journal of Human Resources 28, 572}592.

Jardin, A., 1983. Restoration and reaction, 1815}1848. Maison des Sciences de l'Homme andCambridge University Press, Cambridge.

Knack, S., Keefer, P., 1997. Does social capital have an economic payo! ? A cross-country investiga-tion. Quarterly Journal of Economics 112, 207}227.

Kremer, M., Sarychev, A., 1998. Why do governments operate schools? Mimeo.La Porta, R., Lopez-de-Silanez, F., Shleifer, A., Vishny, R.W., 1997. Trust in large organizations.

American Economic Review 87, 333}338.Lott, J., 1990. An explanation for public provision of schooling: The importance of indoctrination.

Journal of Law and Economics 33, 199}231.Lucas Jr., R.E., 1988. On the mechanics of economic development. Journal of Monetary Economics

22, 3}42.Mankiw, N.G., Romer, D., Weil, D.N., 1992. A contribution to the empirics of economic growth.

Quarterly Journal of Economics 107, 407}437.Romer, P.M., 1986. Increasing returns and long-run growth. Journal of Political Economy 94,

1002}1037.Saint-Paul, G., Verdier, T., 1993. Education, democracy and growth. Journal of Development

Economics 42, 399}407.Steinsaltz, A., 1989. The Talmud: A Reference Guide. Random House, New York.Temple, J., Johnson, P.A., 1998. Social capability and economic growth. Quarterly Journal of

Economics 113, 965}990.Thomas, R.M., Leong, G.K., Mosbergen, R.W., 1980. Singapore. In: Postlethwaite, T.N., Thomas,

R.M. (Eds.), Schooling in the Asean Region. Pergamon Press, Oxford, pp. 184}223.Tornell, A., Lane, P.R., 1999. The voracity e!ect. American Economic Review 89, 22}46.


Documents

Human capital, social capital, and public schooling