Morality matters for economic performance

Guido Tabellini from VoxEU:

Economic backwardness is typically associated with a wide range of institutional, organisational and government failures – and these along many dimensions. In numerous poor or stagnating countries, politicians are ineffective and corrupt, public goods are under-provided and public policies confer rents to privileged élites, law enforcement is inadequate, and moral hazard is widespread inside public and private organisations. There is not just one institutional failure. Typically, the countries or regions that fail in one dimension also fail in many other aspects of collective behaviour.
An influential body of research in economic history, political economics and macroeconomics has shown that both economic and institutional backwardness are often a by-product of history – appearing in countries or regions that were ruled centuries ago by despotic governments, or where powerful élites exploited uneducated peasants or slaves (North 1981, Acemoglu, Johnson and Robinson 2001). But what is the mechanism through which distant political and economic history shapes the functioning of current institutions?
In a recent working paper (Tabellini 2007), I argue that to answer this question we have to look beyond pure economic incentives, and think about other factors motivating individual behaviour. One of these factors is morality. Conceptions of what is right or wrong, and of how one ought to behave in specific circumstances, exert a strong influence on behavioural aspects that directly affect economic outcomes. The list included voters' demands and expectations, citizens' participation in group activities, the extent of moral hazard inside public organisations, and the willingness of individuals to provide public goods.

Values also evolve slowly over time, as they are largely inherited from previous generations. Thus morality, defined as individual values and convictions about the scope of application of norms of good conduct, is an important channel through which distant political history can influence the functioning of current institutions.

Is there any evidence supporting this idea? Exploiting attitudes revealed by opinion polls in The World Value Surveys (Inglehart et. alii 2000), I seek to capture a distinction between values consistent with "generalised" vs "limited" morality. Conceptually, the distinction concerns the scope of application of norms of good conduct (whether towards everybody or just in a narrow group with which the individual identifies). Generalised morality means that individual values support a generalised application of norms of good conduct in a society of abstract individuals entitled to specific rights.
To measure the diffusion of norms of generalised vs limited-morality at an aggregate level, I rely on two variables.

The first is generalised trust towards others (Trust).
This variable has been extensively used in many studies, with two alternative
interpretations – a) as belief about the behaviour of others, and b)as an indicator of moral values and trustworthiness.
The two interpretations are not mutually exclusive. Beliefs are likely to be extrapolated to others’ normative conceptions of how one ought to behave. Moreover, experimental evidence supports this additional interpretation of Trust as an indicator of individual values.

The second variable seeks to measure the values transmitted from parents to children, exploiting a question that asks about the qualities that are appreciated in children. As an indicator of diffusion of generalised morality, I take the percentage of respondents who appreciate respect and tolerance for others in children (Respect). To reduce the scope of measurement error, I consider the average of these two variables, Trust and Respect.

In the paper, I consider evidence from a variety of sources and present two main findings.

First, distant political institutions left a mark in current values, as measured by Trust and Respect. This is evident from individual data on 2nd generation US citizens. Descendents of immigrants from countries that over a century ago were ruled by more democratic political institutions are today more likely to display generalised trust. It is also confirmed by aggregate data on European regions. Within European countries, Trust and Respect are more widespread in regions where centuries ago executive powers were constrained by the prerogatives of independent judiciaries, or by a Chamber of political representatives. Although the precise mechanism of cultural transmission remains to be pinned down, the inference that political history influences current attitudes and values is robust and not dependent on controversial identifying assumptions.

The second finding concerns the contemporaneous link between values and institutional or economic outcomes. This link emerges from a variety of samples. Aggregate cross-country data reveal that countries where generalised morality is more widespread have better governance indicators and specialise in sectors that rely on well-functioning legal institutions. European data also shows that regions with more Trust and Respect are more developed today, and have grown faster since the mid 1970s. Within Italy, voters in regions with higher indicators of generalised morality are also more willing to punish political incumbents who misbehaved.

Since values are endogenous, causality here can be inferred only under additional identifying assumptions. The method of identification exploits the slow-moving component of values. In the cross-country data, I use grammatical rules of the language spoken in the country as an instrument for values. The idea is that language evolves very slowly and its grammatical structure reflects distant ideas and cultural traditions that no longer matter for the functioning of institutions except through current values. In particular, two rules governing the use of pronouns in spoken conversations are correlated with current cultural traits as measured by Trust and Respect. The correlation is strong both within multilingual countries and across countries speaking different languages.
I then consider the component of current values explained by these two grammatical rules, and show that it is robustly correlated with indicators of the current quality of government, and with patterns of specialisation in production. If these grammatical rules are not also directly correlated with other unobserved determinants of the dependent variables, this is evidence of a causal effect of values on governance outcomes and on comparative advantages in specific sectors.
In the within-country data, I exploit a similar methodology, except that I use distant political history in the region as an instrument for current regional values. These regions have been part of a unitary state for over 150 years, but before unification they were governed by very different political institutions. Identification assumes that, after taking into account indicators of economic development at the time of unification, distant political history matters for economic performance only through regional values. The component of current regional values explained by distant political history is strongly correlated with current economic performance in the region. Under the identifying assumption, this is evidence of a causal effect of values on economic development.
These empirical findings do not precisely pin down the mechanism through which values influence economic and institutional outcomes. There are several possible channels of influence, such as the functioning of political institutions, including local governments, or the organisation of production, or the willingness of citizens to respect the law. In fact, all these channels could be relevant, and the working paper presents evidence that, in Italian regions with better values or with a tradition of political independence, voters are more willing to punish incumbent politicians accused of criminal behaviour.

This line of research points to an exciting new research agenda.
Three questions in particular seem important.
First question: how do individual values influence institutional outcomes? In principle, this can happen in several ways, through bureaucratic behaviour, through voters' behaviour, or by making citizens more or less law abiding. Which of these alternative channels is more important in practice?
Second, how do values evolve over time? Why do current values reflect the functioning of political institutions in the distant past? What is the precise mechanism of cultural transmission for values supporting generalised morality? Does it take place within the family, or in other environments? Does it reflect purposeful deliberation or is it an unintended by-product of other activities?
But perhaps the most important question of all concerns the policy implications of these findings. What policy instruments are more appropriate to an environment with adverse cultural traits, and which policy interventions ought to be avoided? Does political decentralisation facilitate the acquisition of favourable values, and is self-government something that can be learnt through experience? And what can be done to facilitate this collective learning process? The promotion of economic development would become much easier if we could answer these questions.