Skip to main content

Trade liberalization, democratization, and technology adoption

Abstract

A general equilibrium theory with heterogeneous skills predicts a complementarity between trade and democracy in creating demand for superior technologies. Trade liberalization or democratization alone may lead to vested interests that limit technology adoption. We use panel data on technology adoption, at a disaggregated level, for the period 1980–2000. Exploiting within-country variation over time and the heterogeneous timing of trade liberalization and democratization, we document a significant and sizable positive interaction between trade openness and democratization for technology adoption. The result that transitions to open democracies are beneficial for technological dynamics is robust to a large set of checks.

This is a preview of subscription content, access via your institution.

Notes

  1. 1.

    See also Melitz and Redding (2014) for a comprehensive overview. Aidt and Gassebner (2010) provide evidence that oligarchic rulers are more free to extract resources in countries protected by trade barriers by, for example, exploiting trade taxes.

  2. 2.

    There is a vast literature studying the determinants of income growth at the cross-country level. Przeworski and Limongi (1993), Barro (1996), Tavares and Wacziarg (2001), Persson (2004), Rodrik and Wacziarg (2005), and Papaioannou and Siourounis (2008) study the effect of democracy, while Greenaway et al. (2002), Dollar and Kraay (2003), and Edwards (1998) focus on the effect of trade liberalization.

  3. 3.

    Aghion et al. (2008) show that democracy fosters value added per worker in the more advanced sectors of an economy by reducing the protection of vested interests and granting freedom of entry into markets.

  4. 4.

    Comin and Hobijn (2009b) and Comin et al. (2012) also highlight the role of political economy and trade in technology adoption by showing how lobbies and geographical distance slow down technology diffusion.

  5. 5.

    Comin and Mestieri (2010) explore the intensive margin of technology adoption further by filtering out the effect of aggregate demand on technology adoption.

  6. 6.

    That the elite do not supply labor is stated only for simplicity. The results require only that these individuals are able to extract resources on top of the returns from supplying labor.

  7. 7.

    This modeling of the production function of the manufacturing sector essentially follows Yeaple (2005).

  8. 8.

    Recall that the price of the Z good is normalized to one.

  9. 9.

    See Acemoglu (2006) for an extensive discussion of these issues.

  10. 10.

    For instance, the assumption that an increase in A affects only the production of good X is only to simplify illustration. The results require only that productivity in the modern sector is relatively more elastic to technological improvements than in the traditional sector.

  11. 11.

    As discussed in Sect. 4, in line with the literature, the empirical coding of the political regime is based on information on the extension of the political franchise (whether it is restricted or universal), on the presence of free and contested elections, and on the extent of substantive political and civil liberties (which are measured by the Freedom House and the Polity Projects).

  12. 12.

    Treating the skill levels as unbounded guarantees an interior solution to Lemma 1 in terms of the threshold level of skills that puts the economy in equilibrium. When skills are bounded, we could have a corner solution in which the manufacturing sector is closed down in an open economy when imports are sufficiently cheap, that is, low \(p^{*}\).

  13. 13.

    Empirically one can explore only the sign, or equivalently the direction, of changes in technology adoption in response to regime changes and not an optimal level of technology. To simplify the derivation of the prediction on the direction of the change we have limited our analysis to the consideration of a costless (marginal) change in A.

  14. 14.

    This appears the most likely scenario in less-developed countries, where the low income workers tend to be politically pivotal; see Tavares (2008).

  15. 15.

    If technological progress is not skill-biased, it always leads to higher real wages for all workers and democratization would always leave the incentives to foster technology improvements unchanged.

  16. 16.

    Notice that, in view of the model presented above, we should not expect technological dynamics to be a main driver of either regime change. The autocrats are not enticed to extend the franchise, which would at any rate not affect (or even dissuade) technology adoption on the part of the new ruler, and do not have any long-run gains from opening to trade, since this would impair their incentives for technology adoption.

  17. 17.

    According to the original classification by Sachs and Warner, a country is defined as being open if none of the following criteria is met: (i) average tariffs exceed 40%, (ii) non-tariff barriers cover more than 40% of trade, (iii) it has a socialist economic system, (iv) the black market premium on the exchange rate exceeds 20%, or (v) there is a state monopoly on major exports. The original index has been subject to several criticisms as it also captures aspects of liberal policies as well as trade policies.

  18. 18.

    The Polity IV variable measures the quality of democratic institution and varies from \(+\,10\) (strongly democratic) to \(-\,10\) (strongly autocratic). To check the robustness of the results we also use alternative codings of political regimes, such as Golder (2005).

  19. 19.

    The conceptualization of trade openness and democracy as dichotomic follows a large empirical literature; see Munck and Verkuilen (2002), Przeworski et al. (2000), and Wacziarg and Welch (2008).

  20. 20.

    For robustness checks we also use information on country-level labor productivity from Mayer et al. (2008) that still offers only an indirect measure of technology adoption at the country level.

  21. 21.

    This level of clustering is chosen as benchmark because the information on trade and political regimes is at the country level. For robustness we have nonetheless also considered clustering the errors at the technology-country level and at the technology level.

  22. 22.

    For many technologies the data reports information on the number of capital goods per capita (e.g., the number of computers per capita). For some technologies the information refers to the output produced (e.g., the amount of steel produced in electric arc furnaces), while for some the data report information on technology level of diffusion (e.g., the number of credit and debit card transactions per capita). We refer to Comin and Hobijn (2009a) for an exhaustive description of the data.

  23. 23.

    This amounts to excluding from the omitted category the countries that are open democracies from the beginning of the sample period. This involves restricting attention to 104 countries from the full sample of 129. We have replicated all the analysis in the full sample (thereby including open democracies in the reference category) and obtained similar qualitative results.

  24. 24.

    For space reasons we directly report the most extensive specifications in columns (5) and (6). The results are similar when controlling for the covariates in the d-i-d framework and when excluding them in the specifications with alternative technology and time fixed effects.

  25. 25.

    See Bertrand et al. (2004) and Giavazzi and Tabellini (2005) for extensive discussions of this issue.

  26. 26.

    For instance, there are about 60 transitions to trade openness (the full list of trade and political regimes and transitions years that is reported in the Supplementary Material). About 40% of these transitions take place in autocracies and 60% take place in democracies.

  27. 27.

    These checks are reported in the Supplementary Material. The findings at yearly frequencies are also robust to the use of alternative codings of democratization and are not driven by a specific set of countries, in particular those belonging to the former Soviet bloc that went through a transition to market economies in the 1990s.

  28. 28.

    The within correlations of the panel with 5-year intervals are 0.03 between demo and openness, 0.36 between democracy and the interaction, and around 0.65 between openness and the interaction.

  29. 29.

    Measures of trade openness often used in the literature, such as the total imports and exports over GDP, cannot be used for our purposes because they inform on the actual trade flows and not on the trade regime changes (or trade reforms). These measures are also typically regarded as likely endogenous to technological dynamics.

  30. 30.

    With non-dichotomous measures the within correlation between openness and democracy, openness and the interaction, and democracy and the interaction are around 0.07, 0.67, and 0.40, respectively.

  31. 31.

    The inclusion of technology \(\times \) year fixed effects can help account for technology-specific dynamic patterns over time. Controlling for the initial level of each technology in each country nonetheless appears a natural, although quite demanding, robustness check.

  32. 32.

    In spite of the different techniques and dependent variable, the findings confirm the insights obtained from GMM results that account for the role of the lag of manufacturing productivity discussed above.

  33. 33.

    Comin and Hobijn (2010) study adoption lags using cross-sectional information on the delay in adoption of each technology in each country so that the unit of observation is at the level of technology-country. In our analysis we exploit variation within countries over time where the unit of observation is instead at the level of technology-country-year.

  34. 34.

    The within correlations between openness and democracy, openness and the interaction, and democracy and the interaction are around 0.05, 0.65, and 0.40, respectively.

References

  1. Acemoglu, D. (2003). Patterns of skill premia. The Review of Economic Studies, 70, 199–230.

    Article  Google Scholar 

  2. Acemoglu, D. (2006). Oligarchic versus democratic societies. Journal of European Economic Association, 6, 1–44.

    Article  Google Scholar 

  3. Acemoglu, D., Aghion, P., & Zilibotti, F. (2006). Distance to frontier, selection, and economic growth. Journal of the European Economic Association, 4(1), 37–74.

    Article  Google Scholar 

  4. Acemoglu, D., Johnson, S., Robinson, J. A., & Yared, P. (2008). Income and democracy. American Economic Review, 98(3), 808–842.

    Article  Google Scholar 

  5. Aidt, T., & Gassebner, M. (2010). Do autocratic states trade less? The World Bank Economic Reivew, 24, 38–76.

    Article  Google Scholar 

  6. Aghion, P., Alesina, A., & Trebbi, F. (2008). Democracy, technology and growth. In: Helpman, E., (Ed.), Institutions and economic performance. (pp. 511–543). Cambridge: Harvard University Press.

  7. Barro, R. J. (1996). Democracy and growth. Journal of Economic Growth, 1(1), 1–27.

    Article  Google Scholar 

  8. Bertrand, M., Duflo, E., & Mullainathan, S. (2004). How much should we trust differences-in-differences estimates? Quarterly Journal of Economics, 119(3), 249–275.

    Article  Google Scholar 

  9. Cervellati, M., Jung, F., Sunde, U., & Vischer, T. (2014). Income and democracy: Comment. American Economic Review, 104(2), 707–719.

    Article  Google Scholar 

  10. Comin, D., & Hobijn, B. (2004). Cross-country technology adoption: Making the theories face the facts. Journal of Monetary Economics, 51, 39–83.

    Article  Google Scholar 

  11. Comin, D., & Hobijn, B. (2009a). “The CHAT Dataset,” Harvard Business School Working Paper, 10.

  12. Comin, D., & Hobijn, B. (2009b). Lobbies and technology diffusion. The Review of Economics and Statistics, 91, 229–244.

    Article  Google Scholar 

  13. Comin, D., & Hobijn, B. (2010). An exploration of technology diffusion. American Economic Review, 100(5), 2031–2059.

    Article  Google Scholar 

  14. Comin, D. A., Dmitriev, M., & Rossi-Hansberg, E. (2012). “The Spatial Diffusion of Technology,” NBER Working Paper, 18534.

  15. Comin, D. A., & Mestieri, M. (2010). An intensive exploration of technology diffusion. NBER Working Paper 16379.

  16. Do, Q.-T., & Levchenko, A. (2009). Trade, inequality, and the political economy of institutions. Journal of Economic Theory, 144(1), 1489–1520.

    Article  Google Scholar 

  17. Dollar, D., & Kraay, A. (2003). Institutions, trade, and growth. Journal of Monetary Economics, 50(1), 133–162.

    Article  Google Scholar 

  18. Edwards, S. (1998). Openness, productivity and growth: What do we really know? Economic Journal, 108, 383–398.

    Article  Google Scholar 

  19. Falkinger, J., & Grossman, V. (2005). Institutions and development: The interaction between trade regime and political system. Journal of Economic Growth, 10, 231–272.

    Article  Google Scholar 

  20. Galor, O., & Moav, O. (2000). Ability-biased technological transition and economic growth. Quarterly Journal of Economics, 115(2), 469–497.

    Article  Google Scholar 

  21. Giavazzi, F., & Tabellini, G. (2005). Economic and political liberalizations. Journal of Monetary Economics, 52(6), 1297–1330.

    Article  Google Scholar 

  22. Golder, M. (2005). Democratic electoral systems around the world. Electoral Studies, 24, 103–121.

    Article  Google Scholar 

  23. Greenaway, D., Morgan, W., & Wright, P. (2002). Trade liberalization and growth in developing countries. Journal of Development Economics, 67, 229–244.

    Article  Google Scholar 

  24. Mayer, T., Paillacard, R., & Zignago, S. (2008). The CEPII trade, production and bilateral protection database, CEPII, www.cepii.fr.

  25. Melitz, M. (2003). The Impact of trade on intra-industry reallocations and aggregate industry productivity. Econometrica, 71, 1695–1725.

    Article  Google Scholar 

  26. Melitz, M., & Ottaviano, G. (2008). Market size, trade, and productivity. Review of Economic Studies, 75(1), 295–316.

    Article  Google Scholar 

  27. Melitz, M., & Redding, S. J. (2014). Heterogeneous firms and trade. In G. Gopinath, E. Helpman, & K. Rogoff (Eds.), Handbook of international economics (Vol. 4). Amsterdam: Elsevier.

    Google Scholar 

  28. Mokyr, J. (1998). The political economy of technological change: Resistance and innovation in economic history. Cheltenham: Edward Elgar.

    Google Scholar 

  29. Munck, G., & Verkuilen, J. (2002). Conceptualizing and measuring democracy: Evaluating alternative indices. Comparative Political Studies, 35(1), 5–34.

    Google Scholar 

  30. Murtin, F., & Wacziarg, R. (2014). The democratic transition. Journal of Economic Growth, 19, 141–181.

    Article  Google Scholar 

  31. Olson, M. (1982). The rise and decline of nations—Economic growth. Stagnation and social rigidities. New Haven: Yale University Press.

    Google Scholar 

  32. Papaioannou, E., & Siourounis, G. (2008). Democratization and growth. Economic Journal, 118, 1520–1551.

    Article  Google Scholar 

  33. Persson, T. (2004). Consequences of constitutions. 2003 Presidential address at the annual EEA congress. Journal of European Economic Association, 2(2–3), 139–161.

    Article  Google Scholar 

  34. Persson, T., & Tabellini, G. (2009). Democratic capital: The nexus of political and economic change. American Economic Journal Macroeconomics, 1, 88–126.

    Article  Google Scholar 

  35. Przeworski, A., Alvarez, M., Cheibub, J. A., & Limongi, F. (2000). Democracy and development: Political institutions and well-being in the World, 1950–1990. New York: Cambridge University Press.

    Book  Google Scholar 

  36. Przeworski, A., & Limongi, F. (1993). Political regimes and economic growth. Journal of Economic Perspectives, 7(1), 51–69.

    Article  Google Scholar 

  37. Rodrik, D., & Wacziarg, R. (2005). Do democratic transitions produce bad economic outcomes? American Economic Review, 95(3), 50–55.

    Article  Google Scholar 

  38. Sachs, J., & Warner, A. (1995). Economic reform and the process of global integration. Brooking Papers on Economic Activity, 1, 1–95.

    Article  Google Scholar 

  39. Segura-Cayuela, R. (2006). Inefficient policies, inefficient institutions and trade. Documentos de Trabajo del Banco de España, 633, 669–685.

    Google Scholar 

  40. Slaughter, M. (2001). Trade liberalization and per capita income convergence: A difference in difference analysis. Journal of International Economics, 55, 203–228.

    Article  Google Scholar 

  41. Stefanadis, C. (2001). Appropriation, property rights institutions and international trade. American Economic Journal: Economic Policy, 2, 148–172.

    Google Scholar 

  42. Tavares, J. (2008). Trade, factor proportions, and political rights. Review of Economics and Statistics, 90, 163–168.

    Article  Google Scholar 

  43. Tavares, J., & Wacziarg, R. (2001). How democracy affects growth. European Economic Review, 45(8), 1341–1378.

    Article  Google Scholar 

  44. Thoenig, M., & Verdier, T. (2003). A theory of defensive skill-biased innovation and globalization. American Economic Review, 93(3), 709–728.

    Article  Google Scholar 

  45. Wacziarg, R., & Welch, K. H. (2003). Trade libearalization and growth: New evidence, NBER Working Paper, 10152.

  46. Wacziarg, R., & Welch, K. H. (2008). Trade liberalization and growth: New evidence. World Bank Economic Review, 22(2), 187–231.

    Article  Google Scholar 

  47. Wood, A. (1995). How trade hurt unskilled workers. The Journal of Economic Perspectives, 9(3), 57–80.

    Article  Google Scholar 

  48. Yeaple, S. (2005). A simple model of firm heterogeneity, international trade, and wages. Journal of International Economics, 65, 1–20.

    Article  Google Scholar 

Download references

Acknowledgements

We are particularly grateful to the editor for guidance during the revisions, to four anonymous referees for useful comments and to an associate editor for very detailed and useful feedback. We also thank Jacques Melitz, Fabian Gouret, and Jose Tavares for insightful suggestions. The paper also benefited from comments from seminar participants at Heriot-Watt, Luxembourg, Reading, the UNCTAD-WTO seminar on trade and development in Geneva, and the audiences at the 7th Annual Conference on Economic Growth and Development in New Delhi, Midwest International Trade Meeting in St Louis, European Trade Study Group in Leuven and the International Workshop on Economics of Global Interactions in Bari. Financial support from MIUR through the project PRIN2015-The Legacy of Institutions for Long Term Development 2015T9FYZZ is gratefully acknowledged.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Matteo Cervellati.

Electronic supplementary material

Appendix

Appendix

Analytical derivations and proofs

Proof of Lemma 2

The Cobb–Douglas preferences in (1) together with equations (2)-(4) imply that the aggregate demand for each good is given by

$$\begin{aligned} p\int _{\underline{\theta }}^{\infty }\theta ^{A}dG(\theta )=\beta E, \end{aligned}$$
(12)

and

$$\begin{aligned} Z(L(\underline{\theta }),N)=\left( 1-\beta \right) E. \end{aligned}$$
(13)

where E denotes total expenditure.

In a closed economy the total demand for each good must be covered by internal production. The product market clears when (12) and (13) jointly hold, that is if, and only if,

$$\begin{aligned} p=\frac{\beta }{1-\beta } \frac{Z(L(\underline{\theta }),N)}{\int _{\underline{\theta }}^{\infty }\theta ^{A}dG(\theta )}. \end{aligned}$$
(14)

Recall that the labor market is in equilibrium at \(\underline{\theta }\) if (5) and (4) jointly hold, which implies

$$\begin{aligned} p=\frac{w(L\left( \underline{\theta }\right) ,N)}{\underline{\theta }^{A}}. \end{aligned}$$
(15)

The product and the labor markets therefore clear at \(\underline{\theta }\) if, and only if, (14) and (15) hold simultaneously, which implies

$$\begin{aligned} Z(L(\underline{\theta }),N)=\frac{1-\beta }{\beta }\frac{w(L\left( \underline{\theta }\right) ,N)}{\underline{\theta }^{A}}\int _{\underline{\theta } }^{\infty }\theta ^{A}dG(\theta ). \end{aligned}$$
(16)

Recall that, for any given \(\underline{\theta }\), the share of workers in the traditional sector is

$$\begin{aligned} L(\underline{\theta }) =\int _{1}^{\underline{\theta }}g(\theta )d\theta =G( \underline{\theta }). \end{aligned}$$
(17)

Given \(w(L(\underline{\theta }), N)L(\underline{\theta }) =\eta Z(L(\underline{\theta }),N)\) from (2), and using (17), the equilibrium condition (16) can be finally expressed as

$$\begin{aligned} G(\underline{\theta })\underline{\theta }^{A}=\eta \frac{1-\beta }{\beta }\int _{ \underline{\theta }}^{\infty }\theta ^{A}dG(\theta ). \end{aligned}$$
(18)

The equilibrium in a closed economy exists, is interior and is unique because:

  • Looking at the limits: (i) When \(\underline{\theta }=1\) the left-hand-side (LHS, henceforth) is zero as \(G(1)=0\). The right-hand-side (RHS, henceforth) instead would be strictly positive as \(\int _{1}^{\infty }\theta ^{A}dG(\theta )>0\). (ii) when \(\underline{\theta }\rightarrow \infty \), the LHS is strictly positive, whereas the RHS goes to zero as \(\int _{{\infty }}^{\infty }\theta ^{A}dG(\theta )=0\). Continuity of the functions insure the existence of at least one equilibrium since the two curves must intersect for a finite level of skill.

  • The LHS of (18) is strictly increasing in \(\underline{\theta }\), while the RHS is strictly decreasing in \(\underline{\theta }\), which guarantees uniqueness.

Notice that this is also the case if the skill level is bounded as the same concept holds. Consider for instance a distribution of skills that is bounded from above at a level \(\tilde{\theta }\). When evaluated in the limit case in which \(\underline{\theta }=\tilde{\theta }\), the LHS of equation (20) equals \(\tilde{\theta }^A>0\), while the RHS is still zero as \(\int _{{\tilde{\theta }}}^{\tilde{\theta }}\theta ^{A}dG(\theta )=0\) \(\square \)

Proof of Lemma 3

Denoting by \(\underline{\theta }\) the equilibrium threshold in a closed economy and rewriting the equilibrium condition (18), define

$$\begin{aligned} F(\underline{\theta },A)=G(\underline{\theta })k-\frac{\int _{\theta }^{\infty }\theta {}^{A}dG(\theta )}{\underline{\theta }^{A}}=0, \end{aligned}$$
(19)

where \(k=\frac{1}{\eta }\frac{\beta }{1-\beta }\). To see the effect of an increase in A on \(\underline{\theta }\) we use the implicit function theorem to get

$$\begin{aligned} \frac{\partial \underline{\theta }\left( A\right) }{\partial A}=-\frac{ \partial F(.)/\partial A}{\partial F(.)/\partial \underline{\theta }}=-\frac{ -\int _{\theta }^{\infty }\theta {}^{A}\left( \ln \theta -\ln \underline{\theta }\right) dG(\theta )/\underline{\theta }^{A}}{G^{\prime }(\theta )k-\frac{ \underline{\theta }^{A}\left( -\underline{\theta }^{A}\right) -A \underline{\theta }^{A-1}\int _{\theta }^{\infty }\theta {}^{A}dG(\theta )}{[\underline{\theta }^{A}]^{2}}}>0, \end{aligned}$$
(20)

because by Leibniz rule

$$\begin{aligned} \frac{\partial \int _{\theta }^{\infty }\theta {}^{A}dG(\theta )}{\partial \underline{\theta } }=-\underline{\theta }^{A}<0 \end{aligned}$$

and because \(\partial L\left( \underline{\theta }\right) /\partial \underline{\theta }>0\).

The observation above also directly implies a reduction in the wage w in the traditional sector following an increase in A. The effect of an increase in A on the skill premium is given by

$$\begin{aligned} \frac{\partial \left( \theta ^{A}/\underline{\theta }\left( A\right) ^{A}\right) }{\partial A}=\left( \theta ^{A}/\underline{\theta }\left( A\right) ^{A}\right) [\ln \theta -\ln \underline{\theta }\left( A\right) \frac{\partial \underline{\theta } \left( A\right) }{\partial A}]. \end{aligned}$$
(21)

From (20) \(\partial \underline{\theta }\left( A\right) /\partial A>0\) and because \(\ln \theta \) is strictly monotonic in \( \theta \), there exists a unique level of \(\theta \equiv \overline{\theta }\left( A\right) \) above which (21) is positive.

The increase in the labor occupied in the traditional sector, Z, implies an increase in the total production in that sector. In principle the equilibrium production in the X sector may increase (due to A) or decrease (due to higher \(\underline{\theta }\)) depending on the sign of \(d\left( \int _{\theta }^{\infty }\theta {}^{A}dG(\theta )\right) /dA\). But the equilibrium condition requires that the positive direct effect of a better technology A dominates and always increases total output in the X sector as well. This can be seen by considering again the condition for the equilibrium in a closed economy,

$$\begin{aligned} G(\underline{\theta })\underline{\theta }^{A}=\eta \frac{1-\beta }{\beta } \int _{\theta }^{\infty }\theta {}^{A}dG(\theta ). \end{aligned}$$
(22)

As shown above, a higher A increases \(\underline{\theta }\), so the LHS of (22) is increasing in A, so that the RHS must also increase, which requires an increase in total production in the X sector: \(d\left( \int _{\theta }^{\infty }\theta ^{A}dG(\theta )\right) /dA>0\). Finally, notice from (15) that the reduction in equilibrium wages w and the increase in the threshold level of skill \(\underline{\theta }\) imply a reduction in p. \(\square \)

Proof of Lemma 4

The equilibrium in an open economy is implicitly characterized by (15) evaluated at the international prices \(p=p^{*}\). That \(\frac{\partial \underline{\theta }^{o }\left( A\right) }{\partial A}<0\) and, therefore, that the labor supply L and total production in Z decrease can be directly verified by looking at (15), (17), and the Cobb–Douglas production function for the traditional sector.

As p is fixed in this case, a higher A cannot be followed by an increase in \(\underline{\theta }\) because this would make the denominator on the RHS increase further while it would necessarily decrease \(w(L\left( \underline{\theta }\right) ,N)\) in the numerator, which would violate the equality. All the remaining results directly follow as in the proof of Lemma 3. One can conclude, since prices are fixed by world markets, that the effect of an increase in A monotonically decreases the utility of the elite and increases that of all workers. \(\square \)

Proof of Proposition 1

We need to characterize the change in attitude towards technological improvements by the part of the political rulers in an autocracy following a process of openness to trade. The total income of the elite, which is the residual claimant of the income produced in the Z sector, is given by \(Y^{E}=(1-\eta )Z\), and by dividing it by the size of elite, denoted by \(\sigma \), one gets the per-capita income of each member of the elite. From (10) each member of the elite strictly gains from an increase in the productivity A if, and only if, their real income increases. From Lemma 3, in a closed economy an increase in A increases the indirect utility of the elite because it increases Z and reduces the price p. From Lemma 4, however, Z decreases in response to higher A in an open economy (while \(p=p^{*}\)). In a closed economy, the elite benefits from technology adoption, whereas this is not true in an open economy. \(\square \)

Table 6 Summary statistics
Table 7 Within correlation matrices: different samples

Proof of Proposition 2

From Lemma 3 the nominal wage of workers in the traditional sector, w(LN), decreases with a higher A in a closed economy. Under autarky, the workers in the traditional sector can gain from technological improvements if, and only if, the reduction of price p more than compensates the reduction in their nominal income. In turn, from (4), the nominal earnings of an individual with skill \(\theta \) working in the modern sector are given by the base wage times the skill premium, \(w(L,N)\left( \theta /\underline{\theta }\right) ^{A}\), that from Lemma 3 can be increasing only for the highly-skilled workers, for whom the increase in skill premium \(\left( \theta /\underline{\theta }\right) ^{A}\) more than compensates the reduction in the base wage, w. Therefore, compared to a closed autocracy, the process of democratization strictly reduces the incentives for technology adoption if the new political ruler (the new pivotal voter) is a worker who loses from technology adoption (e.g., a less-skilled worker). The sign of the effect of technology adoption is unchanged if the new political ruler gains from technology adoption (e.g., a highly-skilled worker). \(\square \)

Proof of Proposition 3

Compared to an open autocracy, the emergence of an open democracy strictly increases the incentives to increase A since, from Lemma 4, in an open economy all workers (the new political rulers) gain from higher A, while the elite lose. Compared to a closed democracy, openness to trade (weakly) increases the incentives for technology adoption since (again from Lemma 4) all workers gain from higher A, while (from Lemma 3) in a closed democracy (only) the highly-skilled workers are (more) likely to gain from higher A. \(\square \)

Summary statistics and partial correlations

We report the summary statistics and the within correlations for the baseline samples of Technology Adoption (CHAT data) (Tables 6, 7).

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Cervellati, M., Naghavi, A. & Toubal, F. Trade liberalization, democratization, and technology adoption. J Econ Growth 23, 145–173 (2018). https://doi.org/10.1007/s10887-018-9155-5

Download citation

Keywords

  • Trade liberalization
  • Democratization
  • Political economy theory
  • Technology adoption
  • Disaggregated panel data
  • Within country variation

JEL Classification

  • F16
  • J24
  • O14
  • P51
  • F59