Article Summary: “Does national culture change as countries develop? Evidence from generational cleavages,”

Summary of JOIE article (First View, 07 October 2019) by Danko Tarabar, Assistant Professor of Economics, College of Business Administration, Winthrop University. The full article is available on the JOIE website.

In recent years, culture underwent somewhat of a renaissance in the comparative development literature, largely owing to Douglass North’s seminal work on institutions as well as the advances in the measurement and availability of culture data. Yet, the idea that culture and economics are closely intertwined is hardly new; for example, Max Weber saw Protestant work ethic as instrumental in the formation of the Spirit of Capitalism, while Karl Marx’s Spectre of Communismarose as a result of perceived changes in the underlying relations of production during the era of industrialization.

What is culture? Among economists, culture is often understood as the socially accepted and expected pattern of values, beliefs, and preferences that encourages certain behaviors/actions and proscribes others. As such, it shapes both the individuals’ maximization rules and constraints. It is perhaps of little surprise, then, that some measured cultural dimensions—independent value orientations—have been found to exhibit statistically and economically significant linkages with socio-economic outcomes; for instance, individualism and trust with greater income and governance quality. The robustness of culture’s impact on economic and institutional development has been confirmed across country samples, econometric specifications, and estimators (see Alesina and Giuliano, 2015 for a review).

While culture exhibits strong temporal stability, it is also seldom static. However, due to its primarily vertical transmission (from parents to children), culture is inertial and slow-moving, and significant cultural change at the aggregate level may be discernible only over longer horizons. Unfortunately, few datasets provide time-varying observations on national culture suitable for cross-country empirical work. An exception is Beugelsdijk et al. (2015), who disaggregate by generation five of the six cross-sectional Hofstede (1980, 2001) indices of national culture: individualism, power distance, uncertainty avoidance, long-term orientation, and indulgence/restraint. Hofstede’s research, according to Alesina and Giuliano (2015: 907), “constitutes by far the most used and cited cultural framework in international business, management, and applied psychology.”

By exploiting the variation in Hofstede dimensions across generations and within countries, we are able to ask (and directly observe) whether increases in material well-being shape the evolution of cultures over time. This question has important implications for understanding both social and economic change, since culture and economic development have long been hypothesized to exhibit a two-way causal relationship. Due to data limitations, however, the channel of influence going from development to cultural change has been understudied in the empirical literature. One exception, on which this paper builds, is Tang and Koveos (2008), who relate countries’ average per capita incomes to the original Hofstede indices, finding statistically significant curvilinear relationships with individualism, power distance, and long-term orientation dimensions. 

Beugelsdijk et al. (2015) reconstruct country-level Hofstede indices from individual-level survey responses obtained from World and European Values Survey data collected over several survey waves between 1981 and 2008. The replicated Hofstede dimensions are then assigned to each of the two non-overlapping age cohorts within countries: the older, born between 1902-1958 (1941 on average), and the younger, born after 1958 (1971 on average). The approximately 30-year gap between the two cohorts lends credence to the idea that the average member of the younger cohort may be considered, on the aggregate, as the direct descendant of the average member of the older cohort.

Theories of cultural change (see, e.g., Cavalli-Sforza and Feldman, 1981; Bisin and Verdier, 2011, 2001) suggest that generational transmission of values is a function of the probability of success of both the vertical (from parents) and horizontal (from society) socialization of children, and that without exposure to the rest of society (or if society is culturally monolithic), children adopt parents’ cultural traits with probability 1, thus precluding any cultural change within a population. A simple stylized representation may be given by: CultchildCultparentf(external) + e, where “external” stands for the impact of non-familial influences on the next generation’s cultural formation with f(0)=0, and eis the random personality attribute. 

Econometrically, I re-state the above as Cultd,i1971– Cultd,i1941Xb + a + ei, where iindexes up to 72 countries and done of five replicated Hofstede dimensions. The independent variable of interest is the “long difference” in real per capita GDP by country, log(GDPpct/ GDPpc1970),[1]which captures the rate of change in the living standards between the approximate year of birth of the average member of the younger cohort and year t. Two end periods for tare chosen in the subsequent empirical analysis: 2010 and 1990. The latter serves to mitigate reverse causation concerns, since cultural values of the average memberof the 1971 cohort were less likely to feed back into economic outcomes during the period of childhood and adolescence in 1970-1990. 

The baseline OLS estimates from cross-sectional regressions indicate that more rapid income growth over the 1970-2010 and 1970-1990 periods is linked to a more pronounced move towards individualistic and egalitarian (lower power distance) attitudes across generations in up to 61 countries. The size of the effect is likewise significant, with a unit increase in  (equivalent to a level increase by a factor of ~2.7 over the specified period) being associated with an increase in Δ(Individualism) and decrease in Δ(Power Distance) amounting to about one-third of their in-sample standard deviations. These results hold when controlling for initial income levels, population size, ethnolinguistic/religious fractionalization, geography (latitude and landlocked), predominant religion, and institutional/historical heritage (Roman or Chinese). 

The intuition behind these findings is that greater economic prosperity better enables younger generations to take command over their own lives and to focus less on survival needs and more on pursuits of goals consistent with personal tastes and preferences, thus empowering personal autonomy and emboldening individualistic attitudes. In addition, development can disrupt existing economic and political hierarchies as increasingly affluent individuals realize their potential and opportunities for mobility, and more assertively demand and influence re-allocation of political and other kinds of power, leading to decreased tolerance of the status quo. In the Hofstede framework, power distance captures the degree to which less powerful individuals in a society accept or even support rigid and unequal distributions of power at school, work, or politics.

The basic conclusion of the results presented in this paper suggests a clear trajectory of national cultures toward greater individualism and egalitarianism (lower power distance) over time. The speed of this shift will likely depend on the rate of population change and generational replacement. This finding may be consequential for understanding comparative development in that it suggests the existence of a self-perpetuating cycle between economic development and certain “pro-development” cultural dimensions. I make two caveats. First, although some dimensions of culture share the overall direction of change, it does not follow that world cultures are converging to a single homogenous type. Earlier research (e.g., Inglehart and Baker, 2000) has argued that common (parallel) trends, rather than common outcomes, dominate the temporal dynamics of world’s cultures. Second, no normative conclusions should be drawn here, and there are no cultural orientations that development “should” promote. Cultural change could likewise entail costs to social cohesion and other economic as well as non-economic outcomes. These and related questions may be examined by future research.

[1]In later specifications, the per-worker form is also used.


Alesina, A. and P. Giuliano (2015), “Culture and Institutions,”Journal of Economic Literature, 53(4): 898–944.

Beugelsdijk, S., R. Maseland and A. Van Hoorn (2015), “Are Scores on Hofstede’s Dimensions of National Culture Stable over Time? A Cohort Analysis,” Global Strategy Journal, 5(3): 223 240.

Bisin, A. and T. Verdier (2001), “The Economics of Cultural Transmission and the Dynamics of Preferences,” Journal of Economic Theory, 97(2): 298–319.

Bisin, A. and T. Verdier (2011), “The Economics of Cultural Transmission and Socialization,” in J. Benhabib, A. Bisin and M. O. Jackson (eds), Handbook of Social Economics(Vol. 1A). North Holland: Elsevier, pp. 339–416.

Cavalli, L. L. and M. W. Feldman (1981), Cultural Transmission and Evolution: A Quantitative Approach, Princeton, NJ: Princeton University Press.

Hofstede, G. (1980), Culture’s Consequences: International Differences in Work-Related Values, Beverly Hills, CA: Sage Publications. 

Hofstede, G. (2001), Culture’s Consequences: Comparing Values, Behaviors, Institutions and Organizations across Nations, Thousand Oaks, CA: Sage Publications.

Inglehart, R. F. and W. E. Baker (2000), “Modernization, Cultural Change and the Persistence of Traditional Values,” American Sociological Review, 65(1): 19–51.

Tang, L. and P. E. Koveos (2008), “A Framework to Update Hofstede’s Cultural Value Indices: Economic Dynamics and Institutional Stability,” Journal of International Business Studies, 39(6): 1045–1063.

Tarabar, D. “Does National Culture Change as Countries Develop? Evidence from Generational Cleavages,” Journal of Institutional Economics, forthcoming, doi:10.1017/S1744137418000280

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.