Skip to main content Skip to navigation

The Impact of Income Inequality and Social Norms on Economic Development

My main research interests are the causes and the aspects related with economic and social development of a country. This encompasses two different areas of research, which are outlined below.

1. Growth and Inequality

I focus on two different but interrelated questions, which can be defined as convergence between and within countries.

1.1 Between Country Inequality

The main purpose of this part of the project is to assess the reasons why countries growth rates and, even more dramatically, per capita income levels differ so widely over time and space, despite the theoretical prediction that some form of convergence should instead occur between countries with respect to both variables. In fact, if on the one hand empirical works agree in denying the hypothesis of convergence in absolute terms, on the other hand the theoretical debate over the last few years has failed to reach a common viewpoint on the subject. The empirical, theoretical and methodological contributions put forward on this issue has originated the so-called ‘convergence controversy’. The idea that has now gathered the widest consensus is that some relative form of convergence between countries can be said to occur, whereas different views still persist on its exact specification. One of such views is the so-called ‘club convergence hypothesis’: according to this, per capita income of countries that are identical in their structural characteristics converge to the same level provided that their initial conditions belong to the basin of attraction of the same steady-state equilibrium. This notion draws on the existence of multiple steady states in the development path of an economy.

The reason for multiple steady states is generally identified with what is defined ‘social increasing returns of scale’; that is, there exists a threshold level in the accumulation process such that once this has been overcome, the economy can benefit from a ‘change in regime’ that improves dramatically its performance. The reason of multiple steady state, however, lies in that if the economy starts with too low a level of the key factor of accumulation, e.g. physical or human capital, then such a turning point ensuring steady and fast growth will never be reached, if not for exogenous positive shocks, thus condemning the economy to stagnatation around what has been called a ‘poverty trap’. In other words, in the face of very bad structural conditions, the accumulation path cannot self-sustain itself toward a high-growth equilibrium. Different accounts of poverty traps have been put forward, differing upon the factor that is deemed as key to accumulation: this has in turn been identified with physical capital, human capital, material wealth, but also the degree of ‘development’ of key institutions, such as financial markets, or those enforcing the respect of contracts and property rights, have been called upon as possible causes of development. Therefore, the threshold level can be seen as determined by these and other ‘structural’ conditions of an economy, and may thus be related with, say, the minimum level of investments required to start an activity, or the minimum level of wealth that can be borrowed in the capital markets, or even the level of tax rates.

What I aim to analyse in my first piece of research is the extent to which reliance on market forces per se helps reducing inequality and foster convergence amongst countries. I deal with a theoretical model of growth that studies a closed economy, but the analysis is easily extendible to the open economy case. Its main result is again one of multiple steady state; that is, an equilibrium of high growth and another of low growth are both possible, and convergence towards one rather than the other is determined by the structural conditions of the economy. However, what is peculiar in the present model is that multiple steady states do not occur because of the existence of some threshold levels, somewhat exogenously determined, in the structure of the economy, as the received literature does. Rather, I explore the possibility that, even without any physically determined ‘barrier to growth’, the economy can all the same fail to reach the highest patterns of growth.

How can that happen? What I analyse is a model in which market 'work well', but not 'infinitely well'; that is, rahter than taking for granted that markets clear, my goal is to study explicitly the dynamical behaviour whereby markets clear. Hence, the model I develop can be thought of as portraying an economy in the process of convergencing towards an equilibrium, rather than simply assuming that this has taken place. More precisely, the three makor theoretical underpinnings of the neo-classical approach are: equilibrium of the markets (i.e. instantenous adjustment of prices); perfect rationality and full information of the agents; and, finally, uniform technical change. I have then replaced these with three different assumptions, which, I believe, are less restrictive and only generalise the argument (in the sense that within my model one can return to neo-classical assumptions when some values of the parameters reach ‘extreme’ levels): therefore, markets are not assumed to be in equilibrium but prices change in accordance with market imbalances; agents do not have perfect information and make mistakes, so that it takes time to adjust to the optimal behaviour; rather than assuming that all the economic sectors are equal in terms of profitability, I have instead explicitly considered two different techniques to which two sectors of the economy are associated, which can have different profit rates. As a result, the behaviour of the economy should be observed, so to speak, in “real time”, rather than assuming that all the process of adjustment to equilibrium has been carried out. A crucial factor of the model is the forces that drive technical change: with respect to this, a key assumption is that technical knowledge can be transferred among firms active in the same sector, thus causing increasing returns to scale at the sector level (evidence can also be presented to support this hypothesis): this is the section in which increasing returns kicks into the model.

This approach is similar to the so-called evolutionary approach, where the economy is described outside equilibrium and this may at best be the long-run upshot of a dynamic process. Some of you will recognize in this the underpinnings of Douglass North’s approach to modelling social phenomena in a dynamical perspective.

The main implication of the model is then that market forces are per se not enough to provide the agents of an economy with the necessary incentives to set the economy on the high-growth equilibrium. Even allowing perfect mobility of the factors of production, i.e. capital and labour force, the initial presence of strongly adverse conditions to high growth, which in the model are identified with a relatively low percentage of skilled labour in overall labour supply, means that, even in presence of a market economy, in which, again, adjustment is not simply instantaneous, the high-growth outcome is not ensured. The intuition is very simple, and can be called one of co-ordination failure. The argument runs as follows: firms do not invest in the relatively high-tech sectors because to do this they would need to pay high wages to the few skilled workers that are available. Moreover, technical progress is also fairly limited in the ‘advanced’ sector, due to the little concentration of economic activities in there. On the other hand, workers do not have any incentive to upgrade their skill as few firms have yet invested in the high-tech sector.

What appears from the reasoning is that a sensible economic policy should first try to tackle the problem of adverse structural conditions; only afterwards can market forces be unleashed to produce efficient outcomes. Instead, in policy makers’ agenda there seems to be a reversal of these conditions: market liberalisation, deregulation, privatisations are thought of as pre-requisites of development, and only afterwards, the benefits of supposed higher growth can be ‘oriented’ to benefit the poor. Also, a sound government finance is often imposed as a conditio sine qua non to mobilise loans from international institutions such as the IMF. Even in this case, in my opinion, the causality should be reversed, as public spending could be needed beforehand to trigger the process of high growth. In my view, international institutions could have another role with respect to reform in developing countries; reform is, in fact, costly, both in polical and economic terms. In particular, even though the shift from the slow-growth to the high-growth equilibrium is potentially beneficial to everyone in the society – it is a Pareto improvement- it can cause some groups to be worst off along the transition path or even in the new equilibrium. If these people have gained power positions, they could block the process of reform. International institutions could then act as ‘guardians’ that the ‘intertemporal social contract’ between the various groups of interests is uphold and adhered to, thus solving possible time-inconsistency problems.

If this model can be relied upon to show what is the long-run behaviour of an economy that only relies on internal forces to grow, the immediate issue now becomes what happens when we have the possibility of foreign direct investments. It is indeed true that backward countries can take advantage of the increase in the capital stock from abroad; however, the question is whether this generates a self-sustaining mechanism of growth or not; the case of Argentina shows us how easily capital can flow into and out a country. Therefore, even though the extension to the case of an open economy would be straightforward, the focus on closed economies is still relevant in order to investigate the possibility of self-sustaining process of growth.

1.2 Within Country Inequality

The second theme that I would like to analyse is that of within country inequality. This is part of a research programme undertaken at the ILO at Geneva under the supervision of Professor Marco Vivarelli, which aims to assess from the empirical point of view the impact of globalisation on income distribution in a sample of 45 developing countries over the last two decades.

That of within-country inequality is probably a less investigated theme than the one before, and possibly it is also less relevant as world income inequality is largely (70%) explained by differences across countries. Though, it has an interest on its own, both because it is an index of social tensions, it may be a target of economic policy, and finally because, contrary to between-country inequality, it is showing an increasing historical trend.

Even in this case, theory is divided as to the impact of globalisation. On the one hand, neo-classical economics emphasises the benefits, in terms of decreasing inequality in developing countries (DCs), stemming from opening up to trade and Foreign Direct Investments (FDI). This prediction is grounded on the cornerstone of mainstream trade theory, namely, the Stolper-Samuelson theorem, which argues that each country will specialise in the production of the good requiring those inputs that the country is relatively more abundant of. For instance, China, which is rich of unskilled labour, will specialise in the production of apparel, whereas the US, where conversely skilled labour is more abundant, will produce PCs. Therefore, the theory predicts that the demand for unskilled labour will increase in DCs as an effect of opening to trade, which implies a reduction of inequality. FDI even strengthen this relation as they will tend to fund the same sectors of the economy in which the country specialises. Another underpinning for this hypothesis comes from the observation that, as happened in the case of the Asian ‘Tigers’, opening to trade passes through a stage of expansion of manufactured exports, which in turn favours the demand for basic educated workers, an intermediate category with respect to that of skilled and unskilled. As an effect, equality would be fostered.

However, a number of theoretical studies lead to opposite implications as to the egalitarian effect of globalisation. First, even remaining within the theoretical framework offered by the Stolper Samuelson theorem, the picture changes radically if the sharp distinction between developed and developing countries is substituted by the more realistic one that admits the existence of more than two groups of countries at similar stages of development. This change, in particular, makes the position of middle-income countries - such as Mexico, Argentina, Brazil - particularly critical, as they have to suffer a worsening of their terms of trade comparative advantage with respect to both high and low -income countries.

Second, similar implications can be drawn if one looks more closely at the content of FDI and the goods traded. As argued by Feenstra and Hanson, the effect of FDI would be to transfer the production of the least skilled labour intensive good from the developed to the DCs. But this good would be relatively skilled intensive in developing countries, thus implying an increase in inequality. For instance, evidence has been provided for the case of Mexico that in recent years the skilled wage premium has in fact increased. This approach is based on the view that there exists a continuum of goods differing as to their skilled labour intensity. If we now turn our attention to the nature of imports of DCs from developed ones, we can observe that openness has fostered the transfer of relatively skilled intensive technologies from developed to developing countries, thus again raising the demand for skilled, rather than unskilled, labour. Other authors find support for a skilled-biased technical change occurring in developing countries as well as in developed ones. Finally, other students stress the relevance of a ‘market stealing effect’ caused by foreign enterprises that ‘crowd out’ domestic activities.

The research project aims to investigate the nature of the relationship, using a more refined methodological approach than those adopted so far (see Vivarelli (2002)). Between the novelties, there is a more comprehensive definition of the concept of ‘globalisation’ with respect to previous works; in fact, this encompasses both the increase in a country volume of exports and imports and the amount of foreign direct investments from abroad, whereas only one of these aspects was generally dealt with in previous literature, and often dummy variables were relied upon to act as indicators of openness. Considering both aspects enables us to capture both the real and the financial link of a country with the rest of the world. Institutional changes will also be included in future specifications of the degree of globalisation.

The empirical analysis should hinge upon a theoretical contribution, whose development I am responsible of. This aims top include in a general framework the variety of theoretical aspects put forward above.

Preliminary results seem to show that, on average, globalisation has a neutral effect on within-country inequality. This may lead us to think that the various theoretical considerations put forward earlier actually cancel out with each other. Further research aims to refine both empirical and theoretical insights in order to acquire a better grip on the results so far obtained. In particular, it will helpful to identify common patterns into the sub-groups of countries in which globalisation did and did not have an inequality-enhancing effect in order to advance novel theoretical accounts.

2. Growth, Technological Change, and Social Norms/Institutions

The second issue in my research agenda is to study the relationship between social norms, technical change and growth. The main idea is that different social systems could be classified as to the way in which they face risk, which is seen as an unavoidable element in every economic and social activity. I focus on three types of such interactions: the relationship between entrepreneur and entrepreneur concerning a ‘technological race’; that between entrepreneur and worker as to the mutual investment to be carried out within the joint productive activity; that between employee and employee as to the ‘insurance’ against unemployment (which can indeed be seen as part of the previous relationship). In the light of the typical economic approach of methodological individualism, these should be intended as stylised relationships shaping the content of institutions related with industrial relations, technological research, and systems of welfare In a very stylised fashion, agents involved in these activities have two strategies available in handling risk, and, correspondingly, two different type of society can be derived: they can decide to put up mechanisms to share risk, or not to share it. This originates, respectively, a co-operative and a competitive set of institutions. Whereas the second form is easily interpreted as a liberist type of society, in which risk is borne by single individuals, the first can either give rise to welfare state systems, or to other type of system where assistance is provided not by the state but by the firm itself [see the case of Japan].

This is only the setting of the model. Now it comes the substantive content: admittedly, this is no more than a working hypothesis: the key idea is that there is a connection, or a good and bad match, between the type of technology and the set of social institutions that become established in a society.

One could wonder if it is sensible to talk about a technology as a whole for the economic system. Indeed, some economists speak about “’national economic systems’, and one can always hope to identify to ‘leading’ as opposed to ‘residual’ sectors in an economy, thus focussing on the key ones and neglecting the others. For ‘type’ of technology, here, I refer in particular to the amount of investments that are needed in order to set up the productive activity: generally speaking, high investments determine high fixed costs and increasing returns to scale, i.e. the average cost diminish as output goes up. This is the notion of static returns to scale; another relevant notion is that of dynamic returns to scale, which refers to the gains in productivity stemming from learning by doing, learning to learn, innovations and all the related issues (appropriability Vs imitation), etc. The main idea here is that the higher the amount of investment that is needed, the higher the ‘cost’ of risk, as clearly the losses in case of unsuccessful business is higher.

I then argue that there exists a twofold relationship between technology and social institutions: on the one hand the type of type generally adopted in an economic system may call for particular types of institutions and influence the current social norms. In particular, more risky economic activity, or, better, economic activities involving higher costs of risk, require risk-insuring co-operative institutions, and vice versa. on the one hand, norms may foster or thwart the adoption of some rather than other technologies. For instance, a “collectivist” type of norm may be at variance with a technology in which individual contribution is perfectly discernible and can then be awarded to induce maximum effort; conversely, an “individualist” norm may not function well with a technology in which individual contribution is not distinguishable from that of others. Overall, co-operative institutions are likely to be better suited with less risky activities, whereas competitive institutions are best suited with more risky activities. In general terms, whenever a mismatch between technologies and institutions takes place, a situation of slow-growth trap, analogous to the one described before, can be said to occur.

I hope this general idea may help to shed some light on the current transformations that are affecting most of the world economies. For instance, the lack of competitive-minded social norms are generally blamed on to explain the partially unsuccessful free market reform in formerly communist countries. Moreover, the current Japan crisis may be interpreted as the upshot of a process in which the economy has moved ‘faster’ than social norms. In fact, the activity share in services is rapidly outstripping that in manufacturing, whereas the costs of the risk involved in the two activities is likely to be different: a working hypothesis is that services activities involve lesser amount of risk. Besides, risk is generally wider for countries at the early stages of development, thus a co-operative set of institutions (rather than a competitive one as advocated by IMF and WB for developing countries) are likely to be optimal.


- Acemoglu, D. and Zilibotti, F., (2001): “Productivity Differences”, Quarterly Journal of Economics, vol. 116, pp. 563-606

- Acemoglu, A. and Zilibotti, F. (1999), “Information Accumulation in Development”, Journal of Economic Growth, vol. 4, pp. 5-38

- Aghion, P., Caroli, E., and GarcíaPeñalosa, C. (1999), “Inequality and economic growth: the perspective of the new growth theories”, Journal of Economic Literature, vol. 37, pp. 1615-60

- Arthur, W. B.: Competing Techniques and Lock-in by historical Events: The Dynamics of Allocation Under Increasing Returns, Stanford: Stanford university Press, 1985

- Atkinson, A. B. and Stiglitz, J. E. (1969), “A New View of Technological Change”, Economic Journal, vol. 79, pp: 573-8

- Azariadis, C. and Drazen A. (1990), “Threshold externalities in Economic Development”, Quarterly Journal of Economics, vol. 105, pp. 501-26

- Basu, S. and Weil, D. N., (1998), “Appropriate Technology and Growth”, Quarterly Journal of Economics, vol. 113, pp. 1025-54

- Benabou, R. (1996), “Equity and Efficiency in Human Capital Investment: The Local Connection”, Review of Economic Studies, vol. 63, pp. 237-64

- Benhabib, J. and Rustichini, A. (1996), “Social Conflict and growth”, Journal of Economic Growth, vol. 1, pp. 125-42

- Bernard, A. B. and Jones, C. I. (1996), “Technology and Convergence”, Economic Journal, vol. 106, pp. 1037-44

- Durlauf, S. N. (1993), “Nonergodic Growth”, Review of Economic Studies, vol. 60, pp. 349-366

- Galor, O. and Weil, D. (1996), “The Gender Gap, Fertility, and Growth”, American Economic Review¸ vol. 86

- Greif, A. (1994), “Cultural Beliefs and the Organization of Society: A Historical and Theoretical Reflection on Collectivist and Individualist Societies”, Journal of Political Economy, vol. 102, pp. 912-950

- Grimalda, G. (2002), “Growth with Competing Technologies and Bounded Rationality”, Southampton University Discussion Papers (forthcoming)

- Grimalda, G. (2001) “A Survey on the Nature, Reasons for Compliance and Emergence of Social Norms”, LIUC Papers, N. 92

- Grossman, S. J. and Hart, O. D., (1986), “The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration”, Journal of Political Economy, vol. 94, pp. 691-719

- Jovanovic, B. and Nyarko, Y., (1996), “Learning by Doing and the Choice of Technology”, Econometrica, vol. 64, pp. 1299-1310,

- Lucas, R. E. (1988), “On the Mechanics of Economic Development”, Journal of Monetary Economics, vol. 22, pp. 3-42

- Nelson R, Winter, S. G. (1982), An Evolutionary Theory of Economic Change, Cambridge, MA: The Belknap Press of Harvard Univesity

- Persson, T. and Tabellini, G. (1994), “Is Inequality Harmful for Growth? Theory and Evidence”, American Economic Review, vol. 84, pp. 600-21

- Silverberg, G. and Verspagen, B. (1994), “Collective learning, innovation and growth in a boundedly rational, evolutionary world”, Journal of Evolutionary Economics, vol. 4, pp. 207 - 226

- Vivarelli, M. (2002). “Trade, FDI and Within-Country Inequality: A Theoretical Note” (mimeo)