IGS- Institutions Government and Society (B) - Second Partial, Dispense di Macroeconomia

Scarica IGS- Institutions Government and Society (B) - Second Partial e più Dispense in PDF di Macroeconomia solo su Docsity! IGS B Institutions and Economic Outcomes CRISTIANA FIRULLO < Formal institutions in the neoclassical growth model: Theoretical framework > The fundamental causes of economic > Growth enhan #@ Formal (and informal) institutions and economie growth > Natural experiments > Informal institutions: theoretical framework and Measurement issues 6 | Interplay accoss (formal and informal) William ]. Baumol, Robert E. Litan, Prof. Carl J. Schramm (2009), Good Michalopoulos, Papaioannou (2014), “National Institutions and Subnational Development in Africa”, Journal of Economics, 129 (e 151-213 Portes& Vickstrom (2011), Diversity. Social Capital, and Cohesion, Anqval Review of ,37:461-79 Hall and Soskice (2011), “Varieties of Capitalism: The Institutional Foundations of Comparative Advantage”, Oxford University Press, cap. 1  As it is shown in this case, in the case in which Agent 1 trusts the other and Agent 2 honours, they will reach the Pareto-efficient allocation of outcomes, which attributes w to both players. However, note that Agent 2 has a dominant strategy, namely, exploit the trust of Agent I, since it will give him the best alternative to the two scenarios (trust and distrust), with respect to the outcomes guaranteed by honouring the other. On the other hand, Agent 1 has no dominant strategy: it would better off by trusting agent 2 whenever he honours, otherwise he would choose to distrust. Since we assume common knowledge of all the elements of the game, Agent 1 will know that Agent 2’s dominant strategy is to exploit his trust and, as a consequence, he will choose to distrust. This leads to the Nash Equilibrium (Distrust, Exploit). It results that in the THG, as well as in other variations of the PD game, rational individuals will choose actions that result in Pareto inferior outcome for the players. There are many different solutions of this Dilemma, in order to motivate the players to cooperate inducing them, eventually, to the strategy profile (trust, honour). Those solutions are alteration of the PD so as to induce the agents to take actions that ultimately result in mutually beneficial outcomes. In other terms, Agent 1 is perfectly aware of the incentive of Agent 2 to be untrustworthy, which result in a lack of incentive of Agent 1 to trust the other. The idea behind this solution is that incentives can be used in order to induce cooperation. Consequently, it is necessary to create incentives for Agent 2 to honour the trust of the other player, in order to reach the desired outcome. There are four different ways to do so. 1) Change the preferences of the players One solution is to change the preferences of the players, through exogenous or endogenous changes to the preferences of the players. In the case of exogenous changes, we can assume that Agent 2 has preferences for honouring trust, which may involve adjusting the perception by Agent 2 of the value of x (<0 rather than >0) or assuming that the marginal utility of x is negative, so that Agent 2 has no longer as a dominant strategy the one to exploit the trust. In the case of endogenous changes, we can introduce emotional pre- disposition for cooperation that generate feelings of guilt for non-cooperation, thus resulting in the internalization of cultural norms which place taboos on the realization of benefits achieved at the expense of the others. Indeed, one view of culture is that it changes the tastes of the individuals directly, so that the resulting behaviour is the desirable one, even in the absence of any explicit penalties. (Lazear, 1995) One example of a model that incorporates exogenous changes to player preferences is the one by Kandel and Lazear, with the introduction of a “peer pressure” function, according to which the pressure felt by one worker depends on its own level of effort, on the effort of his peers and on other actions that he and his pairs may take. They introduced this function in order to induce feelings of guilt and shame, in order to induce higher levels of equilibrium effort by workers. Another way to incorporate exogenous changes to player preferences is the one suggested by Børen, by modifying the utility function, so that to include, not only the wealth and effort, but also a variable which captures preferences for honest and dishonest uses of private information. An example of an endogenous formulation is proposed by Lazear, in which preferences evolve so as to promote cooperation in a genetic model of corporate culture. In this models, there are two types of workers: type A, who are more productive in structured environments, and type B, more productive in unstructured environments. The solution is to make investments in order to increase the proportion of workers with preferences for hard work and cooperation within a structured environment. 2) Write an explicit contract Another way to solve the dilemma inherent in the THG is to write an explicit and binding contract, enforced by a third party, requiring participants to select strategy pair (trust, honour). These contracts take one of two forms: (1) monitoring with punishment; (2) monitoring with incentives. In the case of monitoring with punishment, the parties can agree that Agent 1 will monitor the choice of Agent 2, perhaps at some cost c, and that Agent 1 will call on the third party to punish Agent 2 for exploitation or to compel cooperation. In this context, c represents the monitoring cost for Agent 1, whereas d represents the punishment Agent 2 incurs for exploitation: this punishment needs to be significant in order to let (trust, honour) to be a Nash Equilibrium, because in that way, “exploit” will no longer be a dominant strategy for Agent 2. In particular, we need that: 𝑤 > 𝑤 + 𝑥 − 𝑑, which is possible if and only if: 𝑑 > 𝑥, namely, whenever the marginal benefit of exploiting the trust is smaller than the marginal cost (punishment). Another possibility is monitoring with incentives, where Agent 1 can offer an incentive, a, to Agent 2 to honour his trust. In this context, Agent 1 has an incentive to trust as long as the cost of monitoring and rewarding performance is not too large, namely when 𝑤 > 𝑐 + 𝑎. At the same time, Agent 2 has an incentive to honour as long as the incentive offered by Agent I is greater than the benefit from exploiting (𝑎 > 𝑥). However, it is important to notice that explicit contracting represents a “second best” solution to the PD problem, in part because monitoring is costly, incentives can be distorted and third-party enforcement requires that agent’s action be observable and verifiable. Moreover, the explicit contracts that create the trust, honour Nash equilibrium, do not negate the fact that (distrust, exploit) is still a Nash equilibrium, creating both commitment and coordination problems for the players. Indeed, in the case in which Agent 2 would choose to exploit, Agent 1 would have the incentive not to enforce the contract. 3) Rely on an implicit social contract Implicit social contracts, unlike the explicit contracts, cannot be enforced by third parties, such as courts. Only the parties to the contract can determine whether the agreement has been violated and only they can enforce the contract. Social enforcement typically involves either the threat to nullify or discontinue the interaction if the other player does not cooperate, or involves social sanctions (e.g. ostracism) against the non-cooperator by the player. In this case, c is the cost of monitoring the choice of Agent 2 and e is the effort Agent 1 expends to inflict the social sanction f on Agent 2 for exploitation. In this context, it is possible to reach the desired Nash equilibrium as long as 𝑓 > 𝑥 and 𝑤 − 𝑐 > 0. Here, again, we have multiple equilibria: (distrust, exploit) and (trust, honour). Consequently, players will also have to form appropriate beliefs regarding the probability that the other player will choose the cooperative strategy. (it requires probabilities) 4) Repeated interaction Another possibility is to repeat the interaction between the two agents, which allows the application of the Folk Theorem for infinitely-repeated games, which asserts that virtually any outcome can be enforced by an equilibrium, so long as the probability of repeating the game is high enough (or, in other terms, as the players are sufficiently patient). In this way, cooperative outcomes can be achieved with an appropriately defined strategy of play, exploitation occurring in the current round. In other words, using repeated interaction as a solution of the dilemma requires the implementation of either implicit or explicit contractual arrangements to facilitate the punishing of agents who abuse the cooperation and trust of the others. This will have success as long as the short-term benefit from deviating the arrangement is smaller than the long-term benefit of complying. However, even repeated interaction has many issues, in fact, it does not work well, in general, in modern economies since the punishment is more credible, the smaller the group within it would be applied. The same as collusion, which is more durable over time, the smaller the number of the firms participating to the agreement. Moreover, modern economies are characterized by “anonymous interaction”, which makes even more difficult to apply the punishment and monitoring system. The aim in economics is to show that without trust no market could function and, for this reason, it is important that institutions and other devices provide an incentive to be trustworthy. According to the assumption of rationality, Agent 1 will be “prudent”, whenever he chooses to trust in the case in which Agent 2 has the incentive to honour, and whenever he chooses to distrust in the case in which he knows that Agent 2 has the incentive to exploit the trust. However, there are cases in which Agent 1 chooses to trust even if he knows that Agent 2 has the incentive to behave opportunistically. This is the case of “trust as hope”, in contrast to “trust as prudence”: the former can also be seen as an irrational strategy, but there is evidence of this kind of behaviour. The resulting Solow residual has two main functions: when it is taken over time, it takes into account the growth of a specific economy; when it is taken across countries, it measures development. It is defined as the change in the outcome level which is not explained by the change in the investments made on the inputs. Growth accounting, US (Jones 2016) The last column reports the levels of the Solow residuals over time for US and provides us a measure of the economic growth in the US economy. Development accounting On the other hand, in this other case, it provides us with a measure of development across countries. The TFP (total factors productivity) is a measure which directly affects growth and it is also described as the Solow residual, which captures everything we do not observe which explains the variation in y which is not produced by the investment in the inputs. For this reason, it has been called the measure of our ignorance, since it highlights how much is difficult for us to establish whether a country is poor or not based on the levels of the resources. Lecture 2 Formal institutions in the neoclassical growth model: a “Theoretical” framework The economic performance of one country may be, in a first instance, explained by “proximate causes”, like capital, labour and productivity. However, in social sciences, it is very interesting to look at the fundamental causes of those proximate causes: in other terms, we wonder what are the fundamental factors capable of explaining differences among countries in endowments of technology and physical and human capital? There are four possible explanations: (1) luck (2) geography (3) culture – which is usually identified with informal institutions - and (4) institutions. Luck explains the situation of having very similar countries experiencing different grows, which usually arises with small differences with long-run consequences and the selection of multiple equilibria, like in the case of poverty traps. Moreover, it can also explain the “big push” of the industrialization process: indeed, whenever a firm’s investment decision depends upon expectations about other firms’ investment choices, we should take into account technological complementarities and aggregate demand externalities, allowing for possible “coordination failures”. This game has two symmetric equilibria: they both make a high investment, or they make a low investment. Indeed, the higher investment makes sense whenever the given firm expects an higher demand, but this is of course affected by the choice of the other firm. Geography refers to different dimensions: climate, natural resources, topography or unhealthy environment (e.g. the presence of malaria). Montesquieu in 1748 claimed for the importance of climate, for example: indeed, since individual choices are made in order to maximize the utility function, we should take into account the effect that climate has on the individual utility function. For example, if we study in a place where climate incentivizes hanging out and having fun, the average level of students’ performance could be expected to be lower with respect to the level of performance observed in places with worse climate conditions, where there are less incentives to go out. Myrdal in 1968 claimed for the importance of agriculture as precondition to industrialization and, hence, there are needed certain agricultural conditions in order to allow the transition from the traditional to the modern sector. Sachs and co-authors claimed, instead, for the so- called burden of diseases, and how they affect the individual choice: e.g., why should I study too much if my life expectation is relatively low? One of the problems of using Geography as one of the factors that impact on economic growth is the so- called reverse of causality’s issue: are those “geographic” conditions that cause certain levels of economic growth or is the latter to cause the conditions? Indeed, Diamond, in 1999, argued that geography cannot “depict the whole picture”, due to the big heterogeneity among countries, problems of reverse causality and the presence of several counter-examples in time-series analysis. Culture is another determinant of economic growth and it creates an important interaction with the institutions. Culture is composed by beliefs, values, preferences that influence economic behaviours. It works throughout two important channels: • the individual utility function, clear, for example, when one individual has to choose between leisure and work, save or spend money, taking into account as important feature is level of adverse selection. • cooperation and trust, which impacts on social equilibria and explains also phenomena like “coordination failures” and other market failures. Indeed, individuals operate in the market and the probability that they will interact with each other increases as the level of trust and cooperation increases. Again, here the idea behind is that the accumulation of physical capital is something that has to be decided by individuals according to their utility function and culture has a large impact on the latter. Culture is thought as the overlapping of different dimensions, social capital and informal institutions, as suggested by North. Moreover, culture is relatively time-invariant, in the sense that it remains relatively stable even in the presence of “shocks”: indeed, there are many counter- examples of how the importance of culture is challenged by rapid changes due to reforms of formal institutions. An example of this “challenge” to culture is displayed by the experiment of North vs. South Korea, where, even if they are characterized by similar growth and geography (which allowed for a “natural experiment”, one is the counterfactual of the other), their different institutional frameworks resulted in different outcomes in terms of economic growth. Institutions have been listed as the fourth element impacting on economic growth. Formal institutions are usually persistent over time even if they can be, in theory, modified by collective action. Those institutions that foster accumulation of factors of production (K, L, H), innovation and entrepreneurial activity, help in promoting growth. The main issue is to identify which are the more relevant institutions and which form they take (what is the reference model). Among the Economic Institutions, namely those that help agents to interact with each other in the marketplace are: • contracting institutions, namely those that regulate contracts among agents and horizontal relationships • property rights institutions, namely institutions that guarantee physical and intellectual property rights with patents and that regulate vertical relationships, in the sense that they are able to provide citizens with insurance against elites, politicians, etc. … The provision of property rights is important in the context of economic growth, especially in technical institutions, since their absence may lead to the expropriation and this would prevent from further investments, which is a threat to economic growth. Other economic institutions are important since that are relevant as they foster economic growth are those who act in terms of fiscal and monetary policy, in order to allow for macroeconomic stability, but they are less import. Also political institutions are important for the economic growth, as long as they protect democracy (paper by Chang). Growth enhancing institutions, chapter 5 Baumol If we want to increase the GDP, we look at the evidence that people should pursue the same objective of maximizing their income and wealth: if there is enough income and wealth, there will be an increase in consumptions, interactions between agents in the market place and larger investment. In order to realize this, there are two alternative strategies: § Increase the size of the “pie” (innovation, entrepreneurship) and taking one’s fair share from the increase § Taking more of the pie, whether or not it grows (redistribution) Until the Industrial Revolution, redistribution was an object pursued overwhelmingly, but, on the other hand, the economic growth experienced by industrial countries in the last two centuries is unparalleled in previous history. Redistribution was indeed favoured in the past for two main reasons: first of all, taking more of the already existing pie was considered, in its violent meaning, as a heroic activity; secondly, in the past, there was little certainty that the fruits of innovation and contribution to production would have accrued preponderantly to the individuals who endeavoured to make those contributions. However, in order to foster economic growth, innovation is obviously the way that leads to immediate advantages and it is of course preferable to redistribution. Here, the main idea is that once economies approach the technological frontier—that is, once their living standards are among the highest in the world—they can remain at or become the frontier only by shedding state guidance and adopting some blend of entrepreneurial and big-firm capitalism. In other terms, all economies in the world need some degree of entrepeneuriship to generate radical innovation, yet they also need effective big firms to refine it and commercialize it on a mass scale. There is no a specific and general recipe for economic growth: the nature of this blend/mixture of entrepreneurial and big-firm capitalism differs among countries, according to historical circumstances and culture. However, it is believed that in order to achieve the blend of entrepreneurial and big-firm capitalism, institutions must satisfy four key conditions: 1) “Easy to Start and Grow a Business” (project of the World Bank) First of all, governments should reduce the costs of innovation: the latter should be relatively convenient, hence, the expected gains should exceed its expected costs. REGISTRATION COSTS In order to encourage the formation of innovative entrepreneurial enterprises, governments should lower the costs of “formality”, namely the business and property registration and ease of hiring and firing workers. This means that, in order to make the activity of starting and growing a commercial enterprise relatively easy and inexpensive, the time and cost required to fill out the necessary applications should be kept to minimum, and the same for the time of approval. This can be done, for example, by conducing those applications and approvals online, as it is in many developed countries. If such a system is not as developed as desired, namely, if the cost and delay in “being formal” become substantial, it would be optimal for entrepreneurs to do business without approvals, opting from “informality”. However, this would be a suboptimal solution in the economic system, according to the research of De Soto: informal firms must operate at a small scale to avoid detection by authorities, especially because they are likely to avoid taxes payment. Moreover, since any property that they may control is not formally registered, informal institutions cannot obtain formal bank credit and, hence, they can expand only as they generate and save This could seem dangerous, but we have also to consider that this will result very informative, especially for individual educational and professional decision making: as the high-skilled salaries seem more attractive, people would decide to specialize more and this creates positive externalities and higher efficiency levels. 4) Keep the incentives of innovation over time A necessary condition to foster growth through innovation is keeping those incentives working over time: in other terms, successful entrepreneurs should be induced to keep innovating rather than turning to rent-seeking to protect themselves from competitors. And, in particular, they should keep innovating disruptive technologies that can quickly and radically change the competitive landscape. In other terms, entrepreneurial economies must ensure that their firms are constantly engaged in a “Red Queen game”, where every player’s success depends on his or her ability to match or exceed the current efforts of rivals, so that each is forced by the others to bid ever higher. If this kind of competition does not occur, winners are likely to be content to rest on their laurel/comfort-zone which makes innovation slow down or even ceases. There are two institutions that would seem essential for this task. ANTITRUST AUTHORITIES They are essential to keep this game because they regulate the market in order to (1) avoid that competitors fix prices (2) not allowing mergers between already dominant firms in concentrated markets (3) preventing dominant firms from abusing of their dominant position. However, even if the regulation of collusion and other agreements has mostly beneficial effects for growth by fostering competition and, hence, innovation, antitrust becomes much less effective when it comes to keeping the true “winners”, those who earn a position of market dominance or even monopoly, in the competitive race. Indeed, it is important to prevent winners in one round of competition from thwarting the next generation of entrepreneurs who threaten the previous winners: in other terms, it is important to allow for entry in the market, since firms are capable to enter the market as long as they can offer innovative products or they can bring more innovative production processes. WELCOMING TRADE AND INVESTMENT There is evidence that competition from imports can prod domestic firms that may be getting lazy to actively participate in Red Queen-like innovation. It is important to take into account that even economies that are considered at the technological frontier can significantly benefit from open borders, to goods, ideas and people. Using openness as a policy instrument works through market forces which increase the level of competition. Apart from these four factors, other factors that are essential for economic success at the frontier are culture, education, macroeconomic stability and democracy. Some countries grow more rapidly than others because their cultures are more conducive to growth and, consequently, the cultural framework defines the characteristics for entrepreneurial success. Giving too much importance to culture could lead to a pessimistic view of policy-making, since in certain context the deterministic nature of citizens of one country determined by their belonging to a particular culture, could make useless the adoption of certain policies in order to get desired results. However, there are many counter-examples of countries turning their economies around in a relatively short period of time. We should conclude that culture is relevant, but the institutional environment also matters, since, as policy are specifically constructed, they are able to have an impact on culture and, consequently, on people’s responses. As societies become more educated, the labour forces more skilled, they should grow faster. In the Solow model, additional skills in societies can be represented by additional labour: one highly skilled worker can be thought as the equivalent of two less skilled workers. In this context, education is considered more as an input and, the more skilled employees, the higher the probability of advances in technology and, on its turn, the higher the demand of high-skilled workers. If there is macroeconomic stability, countries will be less susceptible to seemingly period financial crises and, without crises, they have a much greater chance of growing. However, it is again a necessary but not sufficient condition to economic growth, since this stability does not assure those results even in the long run, whereas the previous four conditions will be more relevant. Another key element is constituted by democracy to some extent: the idea is that political freedom is likely to enhance economic freedom and to avoid the possibility that an autocratic leader changes his mind about what is the best for one economy and possibly removes rules favouring entrepreneurship. There is enough evidence that democratic ruled countries, even those that are less developed, tend to grow more rapidly than autocracies. On the other direction, as incomes grow, so does a country’s middle class which is likely to demand for political freedom, whereas already democratic countries are likely to backslide to autocracy when their economies perform poorly. To conclude: One aspect of Doing Business has remained unchanged: its focus on promoting regulatory reform that strengthens the ability of the private sector to create jobs, lift people out of poverty and create more opportunities for the economy to prosper. The notion that the private sector has substantial economic, social and development impact is now universally recognized. Policy reforms catalyse private investment. Promoting a well-functioning private sector is a major undertaking for any government. It requires long-term policies of removing administrative barriers and strengthening laws that promote entrepreneurship. This leads to a “consensus” about the principal ingredients for economic growth: internal liberalization, which promotes an ease of doing and growing business, remuneration of entrepreneurial activity, by the provision of effective patent’s protection and property rights, provision of disincentives to unproductive activities by reducing redistribution and increasing, on the other hand, the incentives to innovate within a more competitive environment, capability of keeping those incentives over time, which involves even international liberalization of goods and capitals and the possibility of firms’ turnover again through a more competitive system. The Washington Consensus set of policy prescriptions has born in 1989 as a list of policies that resulted from a survey conducted by John Williamson on economist and policy experts in Washington DC, who were asked to mention which policies they thought would have contribute most to grow in Latin America. This list has evolved over time in the form of a recipe for economic growth and financial stability such that the world’s major financial institutions, in particular the International Monetary Fund, imposed on a number of developing countries that required bridge financing to enable them to avoid financial crises. The main reason why this list has been thought to be a generalizable recipe for economic growth was that many of the policies mentioned – like fiscal policies, open market, competitive exchange rates - were already present in the developed economies and, as a consequence, they would have worked even in the developing economies. The Washington Consensus is characterized by two key-elements: first of all, it refers to everywhere in Latin America, with an important change of perspective, giving prescriptions for a region instead for a specific country; secondly, it characterized by specific rules, that are the 10 recommendations. This list of prescriptions can be summarized by three specific rules: 1) liberalize in internal and international market and in the labour market (we need flexibility in the labour market) 2) privatize public firms 3) stabilize balanced budget In this context, it is important to make a clear distinction between the short term, where the main driving force is the demand in the market, and the long run, where the key-element is given by supply factors and the natural position of one economy. The Washington Consensus had a huge world-wide influence, especially due to the fact that common beliefs of policymakers and investors are mutually reinforcing, as pointed out by Krugman, who described this impact of the Washington Consensus as a sort of “speculative bubble”. He described this as a political economy cycle, where governments are persuaded to adopt it as markets will reward them, and in which markets will supply so much capital because they think they saw an unstoppable move towards policy reforms. In other terms, the positive results of applying the Washington Consensus are likely to suffer from bias due to the existence of positive expectations; then, the positive result, even if biased, would lead authorities to confirm its validity and this will convince other policy-makers to adopt it. There are both “benevolent” explanations and radical critiques to the disappointing effects of the Washington Consensus. Among the benevolent explanations, one could argue that this list is not complete, in the sense that there is a lack of reforms which make the application of the incomplete list ineffective. One solution is represented by the Augmented Washington Consensus, starting from the rationale that the proposal was correct but incomplete, which causes weakness on the prescriptions. Another benevolent explanation is that, given the fact that the list is correct, the countries have not correctly applied it, due to corruption, missing reforms of some markets and other inefficiencies. Among the radical critiques, there is the inadequacy of “one-size-fits-all” strategies, given the fact that an “intellectual apartheid” actually exists between countries and, hence, it is not possible to generalize this list as a recipe for all developing economies. As Rodrik pointed out, a generalized recipe would constitute a “second best” for one specific economy: fixing n-k problems does not guarantee that we will reach unconstrained maximum of the objective function, it is not guaranteed that economic performance will improve. Each country should be thought to have its own objective function to maximize, hence, maximizing an aggregate objective function would lead to ignore country-specific set of constraints and wife has it. Basically, all countries manage to do this by introducing compulsory maternal period for husbands (even in Italy). This produces a jump in the equality indicator without any real effect in the labour market. De facto indicators have the advantage of evaluating the perception of institutions from the point of view of the most relevant stakeholders, like voters and firms, who typically take actions based on their views. In other words, outcome-based indicators are capable to provide direct information on the de facto outcome of how the de jure rules are actually implemented. However, even those indicators have some limits. First of all, the more those indicators are general, the more difficult is to provide a link between the specific policy interventions and the actual outcomes of interest. Another difficulty has to do with interpreting the units in which outcomes are measured, since they are often measured with arbitrary scales. For example, it could be constructed an outcome based indicator by the results of a survey that asks respondents the rate the quality of public services on a 5-point scale, without providing a real definition of the distinction between the possible scores. Since the construction of those scales will be quite specific from one study to another, across different indicators, it would be more difficult to make comparison between countries: indeed, one country may be characterized by dynamic information, rather than another one and, consequently, one country’s indicator can be more biased than the indicator in another country. The measurement problems are inherent and they cannot be avoided, since they depend too much on the subjective point of view of the respondent, like in self-reported health conditions. The sources of outcome-based indicators can be given by experts or broad samples made by individual, firms and so on. It is difficult to say whether it is better to construct indicators based on the responses of the first or the latter. Expert assessments have several major advantages which account for their preponderance among various types of governance indicators. One is simply cost: it is for example much less expensive to ask a selection of country economists at the World Bank to provide responses to a questionnaire on governance as part of the CPIA process than it is to carry out representative surveys of firms or households in a hundred or more countries. A second straightforward advantage is that expert assessments can more readily be tailored towards cross-country comparability: many organizations have fairly elaborate benchmarking systems to ensure that scores are comparable across countries. And finally, for certain aspects of governance, experts simply are the natural respondent for the type of information being sought. Consider for example the Open Budget Index's detailed questionnaire regarding national budget processes, the particulars of which are not the sort of common knowledge that survey data can easily collect. Expert assessments nevertheless have several important limitations. A basic one is heterogeneity across experts’ evaluations, namely that, just as is the case among survey respondents, different experts may well have different views about similar aspects of governance, which suggests that we should be cautious about relying overly on any one set of expert assessments. Moreover, that the country ratings assigned by different groups of experts are too highly correlated. The point here is a simple one. Suppose that one set of experts "does their homework" and comes up with an assessment of governance for a set of countries based on their own independent research, but a second set of experts simply reproduces the assessments of the first. In this case, the high correlation of two expert assessments cannot be interpreted as evidence of their accuracy, but, instead, it would reflect the fact that the two sources make correlated errors in the measurement. In addition, it has been argued that expert assessments are subjected to a certain degree of bias, in particular, towards the views of the business community: while the public good demands reasonable taxation and appropriate regulation, business people usually prefer low taxes and less regulation. However, this limit is not really compelling: If this is true, then the responses of commercial risk rating agencies who serve mostly business clients, or the views of firms themselves, to questions about governance should not be very correlated with ratings provided respondents who are more likely to sympathize with the common good, such as individuals, NGOs, or public sector organizations. Indicators derived from surveys of broad samples of individuals and firms have the advantage of eliciting the views of the actual beneficiaries of good governance, citizens and firms in a given country. The views of these stakeholders matter because they are likely to act on those views. If firms or individuals believe that the courts and the police are corrupt, they are unlikely to try to use their services. Moreover, these indicators have more domestic political credibility, namely, they are more accepted by political national actors, since it is hard for governments to dismiss the views of their own citizens, or of firms operating in their country. However, those indicators have some limits. First of all, there are the typical issues related to the usage of survey data like the problems in the sampling methodology and the one of the non-response bias. In this context, however, it should be recognized that broad samples would be more representative than expert assessments, since they rely on a wider group of respondents. In general, one disadvantage is related to the way in which the survey is dispensed: survey questions may be vague and open to interpretation, or respondents may be fearful to sincerely respond to the questions, in the case in which surveys are carried out in authoritarian countries. Finally, another issue is represented by the biases due to cultural traits in cross-countries surveys, since culture and related social norms matter for the interpretation of responses. The regulation of entry, Djankov et al.2002 This paper, in the context of the “ease of doing business” (DB) project of the World Bank, whose aim is to quantify the burden coming from the required number of procedures, time and costs of doing business, shows that: countries with heavier regulation of entry have higher corruption and larger unofficial economies, but no better quality of public or private good, whereas more democratic and limited governments have lighter regulation to entry. The provided evidence results inconsistent wit the public interest theory of regulation (Pigou, 1938): according to this theory, unregulated markets exhibit frequent failures, ranging from monopoly power to externalities – which are capable to transform a process which is in principle correct to a negative model - (like pollution: the more the pollution, the smaller the production, the smaller the growth). Consequently, a government that pursues social efficiency should counter these failures and protect the public interest through regulation: in other terms, governments should be able to screen potential entrants, in order to avoid negative externalities (like pollution, bad quality of products, reliability…) and make consumers to buy from “desirable” firms and not by fly-by-night market operators (firms which are not capable to gain trust in the market). According to the public interest theory of regulation, the stricter the regulation, namely the higher the number of procedures that are required, the superior the level of social outcomes. According to the public choice theory of entry regulation, by Tullock (1967), on the other hand, the government is actually less benign and regulation is socially inefficient. Indeed, according to Stigler, regulation is introduced to favour private interests, since it is acquired by industry and it is specifically designed and operated for its benefit. Indeed, incumbent firms are capable to acquire regulations that create rents for themselves, since they typically face lower information and organization costs than consumers. Consequently, regulation makes incumbent firms realize higher profits, by enhancing their market power and lowering consumers’ welfare. Another strand is the one by De Soto (1990) and others, according to which regulation is pursued for the benefit of politicians and bureaucrats: politicians benefit from regulations since it favours consumers and, therefore, it attracts more votes and consensus for their approval. The public choice theory, in contrast, predicts that stricter regulation will be associated with higher corruption and less competition. If the public interest theory of regulation holds, namely, if the regulation of entry serves public interest, there should be a positive correlation between the regulation level and products’ quality, consumers’ health, environmental protection (fewer damaging externalities) and competition; there should be also a negative association between regulation level and corruption and shadow economies. Moreover, if the theory holds, the regulation level should be higher in those countries where government’s interests are more aligned with those of consumers, namely, with tighter constraints to the executive, more political rights and ease access to political power. In other terms, governments should regulate more strictly ceteris paribus. In contrast, the public choice theory predicts that governments which are least subject to popular oversight would pursue the strictest regulation in order to benefit themselves and, possibly, incumbent firms. Their analysis involved a dataset of start-up companies in 85 countries in 1999 and they recorded all procedures officially required in order to obtain necessary permits and to notify and file with all requisite authorities. They also calculated the official time and costs for the completion of each procedure under normal circumstances, assuming that information is readily available and all governmental bodies function efficiently without corruption. Since entry regulation varies significantly across firms, industries and countries, they selected a “standardized firm”. They run three separated OLS regressions. Firstly, they regressed 7 different variables for “public goods” on the log of the number of necessary procedures and on the log of the GDP per capita. They wanted to show whether a stricter regulation is associated or not with higher qualities of public goods. They found no significant increase in the quality of public goods as a result of an increase on number of procedures (stricter regulation). Paper by Acemoglu (2001): The Colonial Origins of Comparative Development Acemoglu and other researchers wanted to investigate the effects on economic performance of institutions by using and Instrumental Variable approach. The idea behind their work is that Europeans adopted very different colonization policies in different colonies, with different associated institutions. In certain colonies, Europeans could not settle and were more likely to set up extractive institutions (exemplified by the Belgian colonization of Congo) that are persistent until present: those places have been characterized by higher mortality rates and scarce rights’ protection with respect to the economic and political power of elites. On the other hand, whenever they found an environment favourable to colonization, characterized by low mortality rates, they have been capable to settle European institutions (also called Neo-Europes) and they were characterized by strong private property rights’ protection and constraints to government and elites’ power. As a consequence, they thought to use mortality rates as an instrument for current institutions, in order to estimate large effects of institutions on income per capita. The rationale is that the colonization strategy was influenced by the feasibility of settlements, in the sense that the disease environment has an impact on the likelihood of establishing European institutions. In other terms, they dealt with the endogeneity of property right institutions: the potential settler mortality rate has an impact on the type of colonization, which influences the institutions settled in the colonies, which affects current institutions and we want to see the effect of the latter on the current economic performance. Potential settler mortality (disease environment) (Z) à Settlements à Early institutions (Xpast) à Current institutions (Xcurrent) à Current economic performance (Y) The focus in not on the settlers’ identity, but, instead, on the conditions in the colonies, because, whenever they could not settle, they created worse institutions. In order to capture institutional differences, they used many variables, but, in particular, they used an index for protection against expropriation, that can assume values between 0 and 10: a value of zero is expected to capture the notion of extractive state, whereas a value of 10 is expected to be appropriate to capture well-enforced property rights. In particular, they used the average value of this index for each country between 1985 and 1995, which is an appropriate measure in order to capture the difference between institutions originating from different types of states and state policies. They used both information on the sample of the whole world and on a restricted sample of 64 countries that are ex-colonies. The Ordinary Least Squares Regression is: The result of the regression shows that there is a strong association between the GDP per capita and and the average protection against expropriation, both in the whole and in the restricted sample. Moreover, they added covariates for latitude and dummies for the African and Asian continents: they found that, although they are significant covariates, the statistical significance of the protection against expropriation remains high. To sum up, this regression shows a strong correlation between institutions and economic performance. However, there are many reasons to think that this correlation is not causal: first of all, there could be a problem of reverse causality, in the sense that it could be the case that richer countries may be able to afford and prefer better institutions, instead of having that better institutions may lead to richer countries; secondly, there could be a problem of omitted variable bias, namely, there may be other determinants of the differences in income levels between countries that are correlated with institutions; finally, since those measures have been constructed ex-post, it is likely that a bias arises leading researchers and analysts to see better institutions in richer countries. These three problems would lead to an upward bias. Moreover, possible measurement error may lead to an attenuation bias in the OLS estimates. To fix this problem of bias, it has been recognized to be reasonable to introduce an instrumental variable approach. First stage regression: they first regressed the Log of Settler Mortality on the Average Expropriation Risk (1985-1995) and it starts from the argument that the mortality rate of settlers is correlated with the type of settlement, which has an impact on the past institutions in the colonies, which is persistent over time. As it is shown, there is a linear relationship between high mortality rates and worse institutions today. They use a RD strategy to identify the causal link between formal institutions and growth, starting from the idea that different institutional contexts can have different unobservable characteristics due to the particular cultures, thus recognizing the endogeneity of institutions. Their methodology exploits in a “quasi- experimental” setting the artificial drawing of African borders that took place in European capitals in the mid-late twentieth century: these borders are completely exogenous to African people since, given the fact that ethnic groups remained in their historical settlement regions, more than 200 African ethnicities resulted to be partitioned across two or more countries: as a consequence, this strategy allows to split the cultural and the institutional effect. In other terms, this identification strategy is capable to take account of differences in both the natural environment and in ethnic-specific cultural and anthropological traits. Taking advantage of this accidental historical event, (taking advantage of the exogenous borders) they compared the economic performance across adjacent regions belonging to the historical homeland of the same ethnic group but falling in different countries and, therefore, subject to different formal institutions. Due to the lack of statistical data in Africa, they used as a proxy for development the level of light density as it is displayed by satellite pictures. The Regression Discontinuity is constructed by using as thresholds that exogenous borders in Africa: if we would observe kind of discontinuity in correspondence of the thresholds such that similar contexts (from a cultural point of view) are characterized by different institutions, the RD design is an appropriate way to isolate the institutional effect. There are two possible estimation methods. The first one is at the country level and it has the following regression: where IQL is a measure for institutions quality in a given country, PD is the log of the population density of a given ethnic group in a certain country, AREA is the log of the land area of the homeland of a given ethnic group, where “i” stands for a specific ethnic group and “c2 for a certain country (determined by the boundaries). In particular, “a” is the variable for fixed effects in a given ethnic group, capturing culture and ethnic-specific features. The dependent variable is the light density for a certain ethnic group in a given country. The second estimation method is at the pixel level and it has the following regression equation: This is a linear probability model which, instead of using as a dependent variable the average level of light density, it uses the exact information of whether the pixel says that it is “bright”: the dependent variable is a dummy variable which takes unitary value if it is bright or zero if it is not. If the values are above the median, we are referring to high institutional levels, if they are below, we are referring to low institutional levels. The other control variables introduced concerned the distance to the capital city, to the sea cost, the surface area, the areas under water like rivers, lakes, and so on. In particular, the distance from the capital city is a very important control variable since it is a good proxy for the enforcement of national institutions: the further one place is from the capital city, the smaller the probability of enforcement of the institutions. Consequently, if one village is too far from the capital city, even if the national institution is relatively good, the enforcement cannot be effective and there would be no beneficial effect on the economy of the village. There are two ways to valid the borders (checking for their exogeneity): there is historical evidence that the partition of the border by the colonizers was completely random and, secondly, seeing the border as the cut-off point in a regression discontinuity design is regressing the observable characteristics of the unity of analysis and observe that there is no discontinuity at the cut-off (the assignment variable has to be continue, whereas the outcome variable should show discontinuity to provide evidence of the impact of being in the treatment group.) There is no evidence of endogeneity. They wanted to investigate whether differences in institutional quality across the border None of the coefficients is statistically significant, which means that there is a smooth transition from one institutional framework to the other, in terms of observable characteristics and this ensures the validity of the identification design. Note that, whenever the fixed effect for “ethnic group” is included in the model, this dramatically reduces the magnitude of the coefficient which also becomes statistically insignificant: consequently, there is evidence of omitted variable bias, by non including fixed effects. In other terms, whenever they do not control for ethnicity fixed effects, they get similar result to the one of Acemoglu, whereas if they control for it they obtain a very different result. The fixed effects in both models are statistically significant at 99% confidence level and most of ethnicity constants are individually significant: this suggests that the African subnational development has a strong local geographical-ecological or/and ethnic specific component. A third estimation method is a Regression Discontinuity Design: where f(BD) is the polynomial of the distance from the border, added because they wanted to identify the effect of national institutions on regional development at the borders; IQL is a discrete dummy variable for the quality of institutions which takes unitary value for pixels falling within countries with relatively better institutions. In Panel A, there are reported the RD results that identify the effect of national institutions on development exactly at the border, by including a common to all partitioned ethnicity RD polynomial (the Global RD polynomial). The coefficient on the high institutional quality quantity is close to zero and statistically insignificant. In Panel B, it is included an ethnic-specific RD polynomial, allowing for the effect of the running variable to be different for each group. be not social capital. This is because social capital is defined independently from its consequences and this causes the possibility of falsify the results. On the other hand, Granovetter claimed that the social interaction has different consequences: the likelihood of success of one individual in society depends on many factors but, in the context of social networks, strong ties (very good friends) have a negative effect in social mobility, whereas weak ties can be helpful in moving to the top of society. In other terms, social interaction and social capital has a positive or negative connotation depending on the object of study, due to the social norms involved in social structures. There are different ways to measure social capital, according to the World Value Surveys: trust, respect for others, obedience, control. One can measure social capital by asking people whether most of people can be trusted or not, but the answer would probably be conditioned by the social environment the individual belongs, if he belongs to a relatively homogenous community or to a heterogeneous one. Respect for others is measured by asking people which among certain qualities are important for children to learn at home and look at the percentage of those that has mentioned “tolerance and respect for others”. Obedience, on the other hand, is the typical value of a hierarchical society, where each individual does not internalize the importance of cooperation. Control (over the individual outcome) is measured according to the extent to which individuals feel free to undertake certain actions or, on the other hand, if they feel influenced by the social constraints. The more they feel free, the higher would be the social capital. The more they feel the burden of obedience, the more hierarchical society, the smaller social capital, probably. In general, measuring values and beliefs is problematic: there are limits in terms of comparability, because of the fixed effects of specific communities/nations/cities/networks, but also according to the definition of those values and beliefs which is still specific. Portes and Vickstrom (2011): Diversity, Social Capital and Cohesion They questioned the claims about the positive effects of social capital and its own definition, in particular, they questioned the claims that immigration reduces social cohesion. They reviewed the literature about social capital by Putnam, Coleman and Bourdieu and they examined the evolution of the concept about ethnoracial diversity, civicness and social cohesion. The three leading questions are: § Is social capital, as it was defined as communitarianism and generalized trust, a powerful causal force? § Is social capital the main basis for cohesion in modern society? § What are the real effect of modern immigration on both diversity and social cohesion? During 1980s and 1990s, Putnam’s work recast social capital as a feature of communities and entire societies. However, a strong group of critics emerged to question both the redefinition of the concept and its consequences. They argued that sociability and participation could have significant downsides. In her analysis of the collapse of the Weimar Republic, the political scientist Sheri Berman concluded, for example, that, although the German civil society was rich and extensive, such that Germany would have provided a fertile soil for a successful democratic experiment, it however succumbed to totalitarianism. In 1998, Portes noted that, in order to take seriously Putnam’s argument, three methodological conditions have to be observed: § Social capital must be defined, conceptually and empirically, as distinct from its alleged consequences; § Measures of social capital must be taken prior to its hypothesized effects to ensure that the causal relationship does not run in the opposite direction; § There must be a control for other variables that could plausibly explain the observed relationship in order to guard against spuriousness. Putnam responded to this argument by providing a vast amount of empirical data and analysing it along lines that conformed broadly to these criteria. The results showed how the Social Capital Index (SCI) related positively to a lot of important collective outcomes like education, children’s welfare and democracy. In particular, he showed that this correlation could have slightly changed when controlling for other covariates but social capital was the single most important explanatory factor. However, the methodological issues highlighted by the authors are: the question of the causal order, the possibility of spurious relationships and the sources of communitarianism and public trust. Causal order The question is whether social capital, as it is captured by the SCI, has that multiple positive causal effects. They considered five key dependent variables that are claimed to be consequence of social capital: child welfare, single parenthood, economic inequality, poverty and general population health. Child welfare is measured by the Kids Count Index, whose item components make it reasonable to believe that they are associated with the civic involvement and trusting attitudes comprising the SCI. However, because the index was measured contemporaneously with the dependent variable, it is not clear which comes first: it is equally likely that associational life and trust lead to lower juvenile delinquency and arrest levels than that the absence of juvenile crime and other forms of deviance promote grater expressions of public trust and social participation. SCI retains a strong positive net effect on the Kids Count Index, but the opposite is also the case. The same thing holds for the regressions of poverty and economic inequality on social capital, where the argument for the reverse in causality is even stronger. Moreover, when we control for other variables, the effects of social capital on poverty and vice versa cease to be significant, suggesting that the original association was spurious. The key methodological issue is that all statistical results proposed by Putnam are based on unlagged correlations where the causal order of variables cannot be established with any degree of certainty: child welfare, low juvenile delinquency and low single parenthood may be part of a single complex representing a better quality of life and determined jointly by the same set of historical factors. This possibility leads logically to the issue of spuriousness. Without including a time-sensitive measure, Putnam’s analysis does not allow us to disentangle the causal effects and truly determine whether or not social capital is causing the positive effects observed. The Spuriousness Question There is the possibility that an alleged causal relationship between two or more variables is due to common antecedent factors and this can be the case whenever the partial association between two variables goes to zero after other variables are controlled, but it could also be due to the situation in which those third variables intervene or mediate a valid causal relationship. In the case of social capital, the prospect of a spurious relationship, instead of a mediated one, is stronger because the argument is couched in terms of a direct positive effect of high levels of social capital on each outcome. A good example is provided by academic performance: Putnam showed evidence supporting the theory that in states that are blessed with a strong associational life and civic citizenry, students do much better in school, providing a strong positive association between these two variables. However, when adding other controls, in particular economic inequality which was measured with the Gini coefficient, the original relationship between the SCI and test scores drops to insignificance. Putnam also argued that there is a strong positive association between SCI and economic prosperity, but he omitted the bivariate associations with other variables. When adding controls, the original relationship between social capital and poverty drops down to insignificance. One important covariate is the lagged economic inequality: indeed, it results logic that inequality, which is a real a hard-to-change structural variable, has the true causal effect on poverty: states that were highly unequal decades ago have much greater relative poverty at present, regardless of how trusting or sociable their citizens happen to be. However, there is exception to this pattern, regarding economic inequality itself: the effect of the SCI on the Gini index does not disappear when we control for other variables, which suggests that civic attitudes and associational life may have an autonomous influence on at least one important outcome. The Origin Question, the determinants of social capital Starting from evidence that here is an intimate relationship between social capital and economic inequality, the logical question is where this social capital comes from and whether it can be produced or recreated in areas where it does not exist. Here, it is shown the claimed strong relationship between social capital and economic inequality, as well as the level of education of a state’s population: both variables jointly account for 22% of the variance in the SCI. If we stop the analysis here, we would have concluded that economic inequality is a key determinant of levels of social capital, but the latter in its turn affects inequality. This interpretation does not take into account the possibility that more basic historical and demographic forces may be at play that affect both inequality and the associational and civic life of communities and states. Indeed, factors like race, regional differences in wealth and in its distribution are basic historical forces that have to do with economic disparities an associational life of citizenry. With the historical variables controlled, economic inequality ceases to have any independent effect on social capital and percent black population becomes the most powerful predictor and this new analysis accounts for 52% of the variation in the SCI. According to the analysis of column II, the non-southern states with a homogenously white and better educated population are those where we find the higher stocks of civic life and community participation, but those places were also those that received the grater waves of European immigration in the nineteenth and early twentieth centuries. One group was highly distinct from others in its settlement patterns and associational life, namely Norwegians, Finns and Icelanders who tended to settle in northern states like Michigan, Minnesota and Wisconsin, and who formed tightly knit, self- sufficient communities where strong egalitarian traditions and participation in collective activities required for survival were the norm. these institutions through established formal channels. Given the downsizes of communitarianism, it is not necessarily the case that Cell B is less preferable than Cell A. To sum up, although communitarianism is an appealing ideal, it cannot provide the basis for the organization of a modern democratic society and, when practiced in excess, it may actually threaten its stability. Indeed, advanced complex nations operate somewhere between Cell A and B, where the overarching cohesion created through organic solidarity is supplemented by manifold forms of association. The role of immigration Although it is reasonable to think that immigration reduces cultural homogeneity and communitarianism, it should not raise serious alarms. High migration moves host societies from cell A to cell B, but the presence of strong institutions averts any risk of systematic breakdown: even if unauthorized migration poses some problems for authorities, the weight of modern institutions is quite sufficient to insure that the flow of newcomers is properly channelled. Indeed, no develop nation in North America or Western Europe has been seriously challenged my mass migration. What instead migration does accomplish is to increase demographic and cultural diversity: their arrival does not actually challenge the class structure which remains the same, but it alters the composition of the working classes. The diversity created by mass migration can actually create positive effects: it is necessary for a good division of the labour in advanced societies. On the other hand, the SCI is correlated instead with more basic processes such as economic inequality and racial segregation, which threaten the long term viability of modern democratic societies, whose solidarity is predicated on the opportunities they offer to all to fulfil their individual goals. Whenever those opportunities are denied to large numbers on the basis of their race or ethnic origin is inimical to higher forms of cohesion based on universalistic and impartial rules. In conclusion, this concept of social capital was the one that prevailed in the public mind and that has been adopted by major institutions such as the World Bank. Is culture (informal institutions) or formal institutions that have a greater influence in economic growth? According to Acemoglu, formal institutions have the greatest effect, and this has some historical examples like Korea, which is divided in South and North Korea with different formal institutions. However, there are several counter- examples, like in the case of regional differences in Italy that prevail the common formal institutions, since the unification in 1861: these cultural differences result in economic differences. Lecture 5: Informal institutions (culture) and economic growth Culture, in terms of social capital, presumably has relevant consequences in economic growth, but they are difficult to be identified empirically due to the fact that there is interrelation among formal institutions, informal institutions and economic outcomes. If we use a traditional approach to identify the partial effect of culture on economic growth, using cross-country regressions with controls, there is the problem of possible endogeneity. Alternative approaches are case studies, analysis with the usage of instrumental variables, difference-in-differences, regression discontinuity design and so on. There is experimental evidence of the impact of culture on economies by focusing on different cultural contexts, but these experiments focus on relatively small samples that are not always representative and, as a consequence, they raise problems of external validity. Another possibility is using an epidemiological approach, which is mainly used in medicine in order to distinguish between genetic predisposition and environmental determinants. The idea is that, by focusing on the incidence of diseases between immigrants and natives, if there results a convergence, this means that genetic factors are less plausible. In economics, the epidemiological approach is used to distinguish between institutions (environmental factors) and cultural factors (beliefs, values, norms, social capital that result to be inherited as well as genetics). The strategy is to isolate the impact of culture in the economy by studying the differences in outcomes between people that share the same institutions but that possibly differ in culture. The idea is that, everywhere, culture is transmitted by parents to sons and cultural differences exists between different countries that immigrants bring with them in the host country. Within the host country, however, economic and formal institutions are homogeneous. The missing convergence of economic and other outcomes between immigrants and natives gives support to the conjecture that cultural factors are relevant. This is the regression to estimate the outcomes in the destination countries, for individuals who come from different (or not) places with their cultures. If the coefficient associated with culture, which is the object of interest, is different from zero, culture has a causal effect on the outcome. However, there is the risk of underestimating the effect of culture, due to the fact that there are other channels of transmission of culture, like school, local institutions and so on: if the individual is a second generation immigrant or more, correlation with culture of parents or ancestors in the country of origin is attenuated. This analysis has the “main identification” assumption: within the host country, individuals are exposed to the same environmental factors, namely, all immigrants and natives face the same (formal and informal) institutional setting. However, this assumption is violated in many contexts, like whenever we have spatial segregation, exclusion in the labour market, in welfare, discrimination, ethnic networks and so on. The epidemiological approach assumes that and focuses on the link In the case in which there are social networks among immigrants: According to recent economic literature, networks across immigrants (and across immigrants and natives) have causal effects on integration (in welfare, in the labour market…) through social norms and information. There is evidence that the causal impact of the culture from a given country of origin that results in the country of destination has an impact on the outcomes through the mediation of the social network of immigrants in the destination country. Evidence also shows that the effect of social norms is probably over-evaluated with respect to information and there are important consequences for policies. Algan and Cahuc (2011): “Inherited Trust and Growth” They wanted to develop a new method to uncover the causal effect of trust on economic growth by focusing on the inherent components of trust and its time variation. They showed that the inherited trust of US immigrants is significantly influenced by the country of origin and the timing of arrival of their forebears. They assumed that the level of trust of people living in a certain period influences the economic performance in that period. Inherited trust in 1935 is the one of second-generation Americans born before 1910, namely, whose parents arrived for certain one generation before 1935, of third-generation Americans born before 1935 and fourth-generation Americans born before 1960. This decomposition excludes any overlap in the inherited trust of the two groups. The table reports the OLS estimates for inherited trust in the period 1935-2000, where Swedish Americans’ group is kept as the reference group: having forebears coming from a country of origin different from Sweden has a statistically significant effect in the level of inherited trust, which turns to be strongly persistent. Correlation between Inherited Trust and Trust in the Home Country If there is a transmission of trust within families, there should be a statistically significant correlation between inherited trust and the trust in the country of origin. Moreover, if there has been an evolution of the trust in the country of origin over the century, the correlation between the trust inherited at the beginning of the twentieth century and the level of trust in the home country should be weaker. In the first column, there are reported the results for year 2000. The correlation between inherited trust in the US and the trust in the home country is statistically significant at 1 percent level. The second column reports the correlation between inherited trust in 1935 and trust in the home country, which results to be no longer statistically significant. This suggest that the inherited trust transmitted in 2000 is different from the one transmitted in 1935. Alternatively, this would suggest that there could have been a convergence in inherited trust of US immigrants as the time spent in the home country increases. R°= 0.19 08 *Swd «DK eNw e Nth 8 0.6 *Fin à £ È E 5 8 ° È 04] «india Ò è 5 ager esp SM 2 cita «Aut £ suk “Ba 5 aczr v «Rus È si *Fra sug È 02 Mx n . svoug ’Afri Pi 0 0.4 -0.3 -0.2 -0.1 0 0.1 Inherited trust in 2000 FIGURE I. CORRELATION BETWEEN TRUST IN THE HOME COUNTRY IN 2000 AND INHERITED TRUST OF DESCENDANTS OF US IMMIGRANTS FOR THE PERIOD 2000 R?°= 0.00 084 *Swd «DK eNw 8 064 ent so I « $ Fin £ Fal € 5 8 © 04- ; ù eIndia a $ «sp «69 Gg Suite o af 2 ita E sui 89 eCzr z pra «Rus È 024 e *Youg *Afri Pt 0- T T T T T T -0.2 -0.4 o 0.1 02 0.9 Inherited trust in 1935 FIGURE 2. CORRELATION BETWEEN TRUST IN THE HoME COUNTRY IN 2000 AND INHERITED TRUST OF DESCENDANTS OF US IMMIGRANTS FOR THE PeRIOD 1935 Carrera: WWE INNA and AEG IOTT_INNA They then focused on the correlation between changes in inherited trust and changes in income per capita over time, measuring changes in inherited trust with the changes in the value of the country of origin fixed effects in the period between 1935 and 2000. They found that the correlation is positive and steady: 45 percent of the change in income per capita is associated with change in inherited trust. Column 1 reports the bottom down estimates without additional control. Change in inherited trust is strongly correlated with change in income per capita. Column 2 controls for changes in initial income per capita. The coefficient associated with inherited trust is lowered but still significant at the 1 percent level. Column 3 checks for potential outliers by excluding Africa. Column 4 controls for political institutions. Column 5 reports the results using an alternative measure of income per capita in 1935 and 2000. To conclude, trust causes growth. The effect results to be economically relevant: GDP per capita in 2000 would have increased by 546 percent in Africa if the level of inherited trust had been the same as the one in Sweden. It would have increase by 69 percent in Russia, 59 percent in Mexico, 17 percent in Italy if they inherited the same level of trust as Sweden. However, we should take into account that segregation linked to level of income affects trust. This paper provides a new empirical strategy to uncover the causal effect of trust on growth. They track changes in trust levels inherited by different generations of Americans from the countries of their immigrant forebears as a measure of the evolution of trust in those source countries. By using this inherited component of trust and its time variation, we are able to isolate the specific impact of trust on economic development relative to other traditional candidates— like institutions and geography—captured by the country fixed effects. Inherited trust turns out to explain a significant share of the economic backwardness of developing countries and an important share of economic differences between developed countries over the twentieth century. R. Fernandez “Does Culture Matter?” In this paper the author reviews the literature about the role of culture in economics, focusing on the epidemiological approach, which exploits the fact that people with different cultures can be subjected to the same institutional framework. Firstly, there is evidence that differences in social attitudes are correlated with cross-country economic outcomes: for example, it has been found that countries where people value thrift are characterized by higher saving rates or, for example, using data from World Value Survey, the percentage of individuals who think that housework is as fulfilling as having a job is negatively statistically correlated with the female labour force participation (LFP) across countries and, lastly, Alesina and Angeletos (2005) have found that countries in which a higher percentage of people is convinced that luck plays a fundamental role in the income process are characterized by higher levels of redistribution, in the sense that an higher percentage of GDP is actually spent in social welfare. Culture can be defined as a body of shared knowledge, understanding and practice: it has been defined as “the integrated pattern of human knowledge, belief, and behaviour that depends upon the capacity for learning and trans- mitting knowledge to succeeding generations;” and “the customary beliefs, social forms, and material traits of a racial, religious, or social group; (and) the set of shared attitudes, values, goals, and practices that characterizes an institution or organization. In economics, individuals are agents that make choices in an economical and institutional environment, according to their beliefs and preferences. In the case in which two societies characterized by identical institutional environment will end up realizing different economic outcomes, an economist would explain those differences by saying that the inhabitant made different choices and they did so because they had different preferences, i.e. different culture. Consequently, differences in culture can be translated in a systematic variation in beliefs and preferences across time, space or social groups. We can think that differences in culture do exist because the choices we are considering are taken within an environment that resembles a game with multiple equilibria: non-identical outcomes (equilibria) will be simply the result of the different strategies of the individuals, reflecting their different beliefs/expectations about the equilibrium outcomes. We can alternatively think that the utility function is not the same for everyone and, consequently, the different strategies are lead by different preferences of the individuals. The speed of cultural change depends on how quickly social beliefs and preferences change over time, which in turn depends on the environment, its opportunities, the interactions between individuals and historical experiences. The basic empirical exercise uses data on individuals that live in one given country but whose parents were born in some other country, the “country of ancestry”. The probability that an individual i from country- of-ancestry c takes some action y_ic is: There are many reasons to question this approach: parents are not the unique transmitters of culture, the relationships with institutions like schools, neighbourhoods and so on will also impact on the individual’s beliefs. Moreover, culture is socially constructed: this means that one behaviour is actually replicated whenever the incentives, i.e. rewards and punishments, are provided by a larger social body. Note also that, with the epidemiological approach, as it is shown in the regression above, we study second generation instead of first-generation immigrants and this has the advantage of avoiding confounding difficulties: for example, first-generation immigrants are more likely to suffer to varying degrees such as the ability to speak in the host country language and the prevalence of ties with non-immigrant family members. Those factors, indeed, will be less important for the second generation of immigrants. However, this also means that the impact of culture from the source country is likely to have been attenuated over time and this would lead to an underestimation of the effect of culture on economic outcomes. Nonetheless, there is evidence of a significant correlation between attitudes in the home country and attitudes expressed by immigrants and their descendants. Fernandez proposed a micro-oriented model of the decision problem of women to join or not the labour market. In extreme synthesis, we would say that a women utility function depends positively on the income of the family, namely the sum of the husband income and the potential income of the wife, and negatively might examine the fate of these new organizations, formally identical, in their diverse social and economic and cultural and political setting. The outcome variable is represented by the performance of institutions that results from the aggregation of twelve indicators like cabinet stability, reform legislation, legislative innovation and so on. This is justified by the fact that there is a kind of overlap between the distribution of social capital, identified by denoting as most civic regions the ones with higher affluence to referenda, and the distribution of free city states. Putnam’s hypothesis is that this social capital depends on the past experience of communes, which affected the civic tradition (measure of trust) at the time of unification, because of the very strong persistence of the cultural level. In other terms, the civic tradition today is affected by the civic tradition yesterday, which comes from these different experiences. However, Putnam has never mentioned the possible endogeneity of social capital. In other terms, Putnam used as an instrumental variable the experience of free city states to capture the effect of civic commitment and social capital on economic performance of the different regions, justified by the high correlation between civic involvement tradition and institutional performance. à By now, there is large empirical literature about the endogeneity of social capital/culture: there is the possibility of reverse causality, omitted variables and measurement errors. However, various solutions are suggested, like the one of using instrumental variables. For example, Guiso et al. (2016) showed that Italian cities that achieved self- government in the Middle Ages have a higher level of civic capital today than similar cities in the same area that did not. The size of this effect increases with the length of the period of independence and its intensity. This effect persists even after accounting for the fact that cities did not become independent randomly. Aghion, Algan, Cahuc, Shleifer (2010) formalized a model in which individuals have to decide whether to be civic or not civic. Non civic individuals generate negative externalities like pollution, for example, whereas civic people generate positive externalities, such as increase in the GDP level. Individuals have also to choose to be entrepreneurs or routine producers (like those who work for the state factory), where the productivity of a civic entrepreneur is said to be greater than the one of an uncivic entrepreneur, which is in its turn greater than the one of the routine producer. In this model, regulation exists in order to avoid coordination failures, in particular, to solve externalities and, besides the two choices of being civic or uncivic and of being entrepreneur or routine producers, all individuals have to vote to choose regulation. Uncivic officials use their power in order to implement the rule to demand a bribe. In other terms, regulation gives the power to produce positive or negative externalities, depending on who are the officials. This complex model has two equilibria: a good and a bad one: § In the good equilibrium, individuals choose to trust each other (they choose to be civic), which means that no negative externality is produces and, therefore, there is no demand for regulation. The lack of demand for regulation means that possible uncivic officials, who are basically self- interested, have not the possibility to demand to ask the bribe. There is no reason to implement any rule because this situation is at the equilibrium. § In the bad equilibrium, individuals do not trust each other, they are uncivic. Consequently, they produce bad externalities, which creates a demand for regulation. In this case, there is the need for rules and officials get the power, but, since they are self-interested, they will catch the opportunity to demand for the bribe, through the implementation of ad hoc rules. In other terms, there is a kind of complementarity between formal and informal institutions, which result to be self-reinforcing: distrust is a lack of civicness, which creates negative externalities, which causes demand for regulation, but also regulation causes distrust. There is empirical support based on correlation and the analysis points to a broad complementarity between trust and free market economics, which remains to be explored. Hall and Soskice wrote a book named “Varieties of Capitalism”, in which they elaborated a new framework for understanding the institutional similarities and differences among developed economies. The varieties of capitalism’s approach to the political economy is actor-centred: it is based on methodological individualism, according to which individuals are self-interested and make choices in order to maximize their “happiness” that is express in the form of a utility function. Consequently, political economy is seen as a terrain dominated by multiple actors that seek to advance their interests in a rational way and in strategic interaction with others. According to their view, in capitalistic economies, the individual/actor to study is the firm and, consequently, this actor-centred view is firm-centred. In other terms, firms are used as unit of analysis as they are the key agents of adjustment in the face of technological change or international competition, whose activities aggregate into overall levels of economic performance. The performance of each firm, according to their view, has a relational conception: the evaluation of one firm relies on the quality of relationships that it is capable to establish both internally and externally. However, firms, in their internal and external relationship, are usually exposed to coordination problems and their success depends substantially on their ability to coordinate effectively with a wide range of actors. This also means that we can compare national political economies by reference to the way in which firms resolve coordination problems. rather than in another, when both are feasible in the presence of a specific set of formal institutions. Actors learn to follow a set of informal rules by virtue of experience with a familiar set of actors and the shared understanding that accumulate from this experience constitute something like common culture. The implication is that the institutions of a nation’s political economy are inextricably bound up with its history in two respects: firstly, they are created by actions that establish formal institutions and their operating procedures and, secondly, they are affected by a set of common expectations that result from repeated historical experience and this “common culture” is the main determinant of effective coordination between actors. In the context of industrial relations, CMEs are characterized by higher protection of the labour, whereas LMEs by a lower one, which makes the workers’ turnover much easier. This has an effect on education and training: indeed, acquiring specific skills results to be convenient as long as labour protection is sufficiently high, otherwise, a worker would prefer to acquire general education and increase the probability of a second employment, in the case in which he will be fired by the firm. In the context of corporate governance, since, in LMEs, firms compete in rapidly changing markets, they would need finance available on publicly assessable information, hence, they should be attentive to share price or current profitability in order to ensure access to finance or deter hostile takeovers. This means that the possibility of financing depends on the short-term profitability of firms. Moreover, it is easier for firms to release labour whenever they face pressure from financial markets, which allows them to take advantage of the shifting market opportunities that often present themselves in economies characterized by highly mobile assets. In contrast, in CMEs, the financial system typically provides companies with access to finance that does not entirely depend on publicly available financial data or current returns. Since the finance is not dependent on balance-sheet criteria, investors should have other methods of monitoring performance, which is represented by the so-called “private information” and the usual counterparts are banks. Reputation within the network has a central importance in CMEs: there are few incentives to provide false information due to the consequences in terms of reputation within the network. Access to finance from banks, for example, allows firms to retain a skilled workforce through economic downturns and to invest in projects with long-term returns. In internal relations, LMEs are characterized by hierarchical decisions, where top manager would have the capacity for unilateral decision, whereas, in CMEs, managers must secure agreement for major decisions from supervisory boards, which include employee representatives and major shareholders, and from other managers, suppliers and customers. The CMEs internal relations’ coordination encourages the sharing of information and the development of reputations for providing reliable information, thereby facilitating network monitoring. In external relations, since firms in CMEs use long-term labour contracts, scientific and technical personnel cannot easily move across companies, as in LMEs happens: firms in CMEs cultivate inter-company relations that facilitate the diffusion of technology across the economy and these relations are supported by a number of institutions. As a consequence, CMEs are characterized by cooperation between other firms in the industry. On the other hand, inter-company relations in LMEs are based on standard market relationships and enforceable formal contracts. Those relationships, in the USA, are mediated by antitrust regulations designed to prevent companies from colluding to control prices or quantities, for example. Consequently, LMEs are characterized by competition, instead of cooperation. However, there is no way to identify the causal effect of one institution on another, because there is big problem of endogeneity in the different institutions, which result to continuously interact with each other. In general, it is not possible to say that one kind of capitalism is better than the other. According to the concept of comparative institutional advantage, the particular institutional structure of a political economy provides firms with the advantages of engaging in specific types of activities: in other terms, firms can produce certain goods more efficiently than another because of the institutional support they receive for those activities and this support is not equally distributed across nations. Most of the literature, however, has focused on market relationships in order to explain efficiency gains for firms, but it has not considered the importance of non-market relations that can actually affect outcomes. The point is that national institutional frameworks provide nations with comparative advantages in their particular activities and products. The most relevant type of comparative advantage is innovation which is crucial for firms’ long-run success. There are two types of innovation: radical innovation, that entails substantial shifts in product lines, development of entirely new goods or major changes to the production processes, and incremental innovation, marked by continuous but small-scale improvements to existing product lines and production processes. CMEs should be better at supporting incremental innovation because of the relational requirements of company endeavours: it will be easier to secure incremental innovation where the workforce is skilled enough to come up with such innovations, to risk suggesting changes to product or processes that may affect their work situation. In general, incremental innovation should be most feasible where corporate organization provides workers with secure employment, autonomy from close monitoring and opportunities to influence the decisions of the firm, because the close inter-firm collaboration encourages clients and suppliers to suggest incremental improvements. Moreover, incremental innovation is very important for maintaining competitiveness in the production of capital goods and, therefore, it has a strict link with the firm’s reputation. By contrast, the institutional features of LMEs tend to limit firms’ capacities for incremental innovation. Financial market arrangements in LMEs tend to emphasize current profitability and top managers enjoy higher power for unilateral decisions, depriving the workforce of the possibility to efficiently cooperate in innovation. On the other hand, the institutional framework of LMEs tends to be highly supportive for radical innovation, since, whenever they want to develop a new product or process, they are capable to hire new personnel adequately skilled for their purposes, as well as capable of releasing them if the project turns out to be unprofitable. Finally, it is probably more complex moving from LMEs towards CMEs, because the latter is constructed centred in culture, social partners and experience, which are elements not easy to achieve in a relatively small period of time. Among the main problem VoCs, there is the fact that, due to the large number of firms’ relations, it is not possible to identify which institutions are the most relevant and what is the interaction and its extent between them. Secondly, due to the endogeneity that is evident from the interactions between different institutions, it is not possible to identify the causal effect of one institution on another one. Moreover, there is an unclear identification of the varieties of capitalism and other kinds of political economies different from these two ideal types are not considered. Indeed, the Italian case, for example, cannot be captured by LMEs or CMEs. Chang, in 2011, claimed that GSIs, namely, Global Standard Institutions, that are usually those that are said to be better than others, are usually those of Anglo-American countries, that are seen to maximize market freedom and protecting private property rights most strongly. Those countries started attaching governance conditionalities to bilateral aids, forcing developing countries to adopt GSIs. GSIs, however, are institutions that favour the rich over the poor, capital over labour, finance capital over industrial capital. Hall (2013) tried to make a classification of various types of capitalism, starting from LMEs and CMEs and adding the Nordic Coordinated Economy and Mediterranean Market economy. The starting idea is that, since cooperation leads to positive externalities and positive economic outcomes, policy makers should encourage individuals to cooperate with each other and the place where cooperation. According to the Consensus Theory, cooperation has to occur within the market and the LMEs was said to be the best model to solve coordination problems. On the other hand, Hall’s approach focuses even in non-market forms of cooperation, with many socio-economic consequences, like inequality in the liberal market economy, or individual risk which characterize CMEs. In contrast with the Consensus theory, among these types there is not a “first best”, in the sense that there is no a single way to solve coordination problems, and no model can be said to be better than another. The absence of a first-best implies that there is no (need of) institutional convergence, because there is no ranking of institutions. If instead there was a “Pareto-superior” institutional model, policy makers would have been expected to make any effort in order to shift to the better model: there would be a need for convergence and this will be expressed by historical events or people that would demand the institutional change, since the current model keeps on performing worse and worse, which makes impossible to reach an equilibrium. Consequently, those different alternatives can be seen as different stable equilibria in each country. The second implication of this view is the impossibility to apply “one-size-fits-all” strategies, since countries differ one from another. Thirdly, it is not possible to consider a strategy the one of transplanting parts of institutional context to other contexts. This is because, even in the case in which one model (like LME) is said to be superior to the others, the transition will remain difficult due to the non-transferability of institutional blocks. Moreover, by partially implementing pieces of an institutional framework, it is not possible to predict the final outcomes. For example, lowering labour protection in Germany does not mean that there will reached same outcomes as in US and in New Zealand. This unpredictability explains the failure of many reforms inspired by the Consensus Theory. • institutions of (economic) scientific production, (discussed in Romer in 2016) that create bias since there are usually not sufficient incentives to publish about unexplored fields. Roamer in 2016 suggests that the sources of this problems are: - coordination in science: indeed, science is a social system that uses competition in order to direct the self interest of the individual to the advantage of the group, but, as in other context, competition is vulnerable to collusion, which represents a coordination problem. - social norms in scientific communities, that result to be evident as long as the probability of being cited decreases whenever the scientific research proposed goes against the leading view, with a sort of price associated with open disagreement. - culture, in the form of some honour code for anyone who criticizes openly a revered authority figure. A norm that places an authority above criticism helps people cooperate as members of a belief field that pursues political, moral or religious objectives. This creates bias in scientific research since tolerance of an obvious error is even more corrosive to science than committed advocacy of error. All these aspects suggest that it results to be very difficult that very innovative papers will be published but also written, since authors know in advance that their works will be less likely to be successful, rather than researches in already touched fields or with known approaches. According to many researchers, the presence of academic and professional economists with serious technical training and backgrounds represented a handicap in many professional contexts (like the Monetary Policy Committee). Economics students have often showed their dissatisfaction with their teaching, characterized by a lack of pragmatism, which is necessary for economic understanding, instead of a deep focus on paradigm. In particular, whereas at the undergraduate level, macroeconomics textbooks are regularly revised in order to take into account important issues of crisis and particular shocks, which allows for a productive debate in class, at the graduate level teaching is still tied to a benchmark neoclassical framework. The problem is that a deep focus on a unique framework makes it difficult to truly understand economic dynamics. There are several initiatives, like the paper Gattopardo Economics, that try to develop ideas that may diverge from mainstream economics. Neoclassical and Neo-Keynesian Macroeconomics Usually students are exposed to the neoclassical framework in economics, which causes a specific focus instead of allowing for further research. Neoclassical and Neo-Keynesian frameworks are characterized by their micro-foundation, namely a basis of theoretical microeconomics that allows for formulating variables such as consumption in a given time. A micro-founded macroeconomics model starts from the optimization problem of choosing the optimal level or labour over time, using as objective function the integral over the infinite periods of time of the utility function multiplied by a discount factor. The main feature of being micro-founded is treating the decision problem is addressed to an individual that turns to represent a collectivity as it is understood as a representative agent. The representative agent acts in the market in order to maximize his own utility function (that should be representative of the collection of many utility functions), but this does not allow to account for inequalities between different economic agents. Moreover, they are characterized by dynamics, in the sense that they consider many periods of time, hence the integral sums an infinite horizon of time, or there may be overlapping generation models. In order to add dynamics in the short run, it is possible to add an ad hoc shock per period t. A model such these is the Solow model augmented by the individual optimization problem: when the value to model is savings, this is called the Ramsey model. An analytical framework is used in order to model dynamics: for example, in the short run, dynamics is given by adding ad hoc shock for the given period of time. Most of the models are characterized by the absence of market failures and by the assumption of complete markets. A market is said to be complete as it is characterized by negligible transaction costs and perfect information: this implies that, in each possible state of the world, there exists a price for every asset in the market. According to the first theorem of welfare economics markets lead to the Pareto-optimal solution. The frictions in the market are introduced with reference to specific sectors, in the context of the problems arising from public goods: patents in R&D, taxes and public expenditure in the public sector, as they create a distortion in the optimal level of consumption of individual, in the labour market minimum wages, EPL and unions: in general regulation is understood as another friction in the market. But the truth is that in R&D, there is the need to protect patents (indeed inventions in technology, as long as they are not protected, they are public goods that can be used by anyone in order to increase profitability, there could be the problem of free-riding and so on), in the public sector there is the need to specific forms for taxes and public expenditures, in the labour market, there is the need to protect workers. The main implication of most of the micro-founded models is that the model of liberal market economy is a superior to other institutional frameworks, since, given perfect markets that allow for high level of competitiveness (assumed almost perfect), there is scarce justification for public intervention, neither in terms of efficiency nor in terms of equity, but it is instead likely to introduce inefficiencies in the economy. Solow, in 2008, claimed that, accidentally or not, folding an imperfection into the Ramsey model is likely to push the policy implications in the laissez-faire direction, because without any intervention, the assumption of perfect markets leads to optimal solutions, in theory. Chang, 2011, “Institutions and economic development: theory, policy and history” According to Chang, there are two main theoretical problems in the study of institutions in order to promote economic development. The first issue concerns causality: the focus is usually on specific institutions that promote economic growth, but there is enough historical evidence that economic development shapes institutional change. This is because of three main reasons: firstly, increased wealth creates higher demand for high-quality institutions; secondly, increased wealth makes those high-quality institutions more affordable; thirdly, economic development creates new agents of change demanding even different institutions. The second issue concerns the fact that the relationship between institutions and economic development is usually theorized in a very simple, linear and static way (as it can be founded using neoclassical approaches). Most of the theories argue that liberal market economies are superior, because assuming perfect information in markets leads to the optimal solution of minimizing public intervention. Consequently, according to these theories, maximizing freedom causes maximizing economic growth. For example, fixed salaries that are usually guaranteed by non liberal market economies, with the intervention that introduces minimum wages, do not provide enough incentives to workers and, thus, it will be less likely to increase firms’ profits and economic development. However, there are theories that show that it is not always the case that institutional structures that give the maximum business freedom are likely the optima from a social point of view. For example, acquisitions can reduce competition by lowering the number of active firms in a given sector and giving monopoly power to one firm. Or, for instance, giving firms the freedom to cumulate individual risk without regard to systematic risk is not good for the overall economy, as the financial crisis in 2008 has shown. Or, as another example, freedom in the labour market would lead to the usage of children labour: firms can benefit from using it in the short run, but that may harm all firms in the long run, by harming children’s health and education and reducing the overall quality of labour force. Besides the social side, it is not always the case that freedom in markets guarantees economic development and market failures are pervasive. To sum up, here are thousands of examples that show that freedom in the market does not lead to procompetitive outcomes in the market. In reality, it is theoretically impossible to have complete markets (actually, “autarky”, the absence of markets, is the norm, not the exception) and, hence, the assumptions of neoclassical macroeconomic models are usually violated. Akerlof and Shiller in “Phishing for Phools” argued that all the markets are affected by asymmetric information. If complete markets (and perfect information) exist, in order to make any of these function, even if they consume a very small amount of resources, the resources of the whole universe will be exhausted. Since the complete market assumption is violated, it becomes tricky to identify good institutions: in particular, if the assumptions are violated, the absence of state intervention is no longer an optimal solution for economic growth. Stiglitz, in 2017, argued that there is a parallel between welfare state and developmental state: the second one recognizes that markets on their own did not succeed in their structural transformation that were required if countries were to achieve their developmental ambitions. In such a context, developmental state corrected these market failures and had a catalytic role in promoting structural transformation. In other terms, whereas welfare state aims at substituting himself to markets, developmental state collaborates with markets to avoid inefficiencies. According to Chang, adopting GSIs (Global Standard Institutions, the same as Consensus Institutions) in developing countries may be unsuitable to promote developments in certain countries or in specific periods of time. First, because of market failures that make the liberal economies sub-optimal in general. Secondly, because of the availability of resources: indeed, Rodrick noticed that each country has its own public budget with specific constraints and the need for balance of payments. Thirdly, there are institutional complementarities that make the outcomes of institutional change unpredictable. Finally, the optimal institution strongly depends on the stage of development of one country and of its firms: late-comers, namely nascent industries usually require high protection before opening to competition, whereas opening to free-trade is usually a very good option for strong industries. There are actually possible interventions to move from left to right and promote economic growth, that not always rely on liberal market economy. Nowadays, the inequality level is more than it was before the Great Depression and before the World War. It is usually very difficult to run a cross-nation analysis, but according to it in the US there was the highest frequency in top 1% incomes pre- taxes. There is another way to define inequality based on how much income/production is addressed to capital and how much to labour. The fact that inequality defined as the Labour Share of Income has some trend during time in some country violates the assumption in the Cobb-Douglas production function that the share of labour and capital is stable over time. There is evidence that inequality has started growing faster with Capitalism, namely when the Capital Share of income keeps growing. Keeping productivity constant, the share of labour on total income, according to the Cobb-Douglas function, should be constant over time. However, it is decreasing over time with the raise of capitalism. The possible causes of inequality are: 1. market forces, namely all those forces that have an impact on the demand and supply of factors. They are: a. immigration, which may lead to an increase in inequality level as it usually causes an increase of labour supply for unskilled workers, like in the US case. From the equilibrium, the wage level of unskilled workers decreases, while the one of skilled workers remains stable or increases. In any case, inequality increases. b. international trade, which allows for import of labour from abroad. If labour abroad is cheaper and there is the possibility for import from abroad, the demand for unskilled labour shifts towards those markets. Differently from immigration, where the change is in the supply side, here is in the demand side. c. skill-bias technological change and “Capital-skills complementarity”. By skill-bias technological change we mean a shift in the production technology that favours skilled over unskilled labour, by increasing its relative productivity and, therefore, its relative demand. Capital-skill complementarity means that the elasticity of substitution between capital stock and unskilled labour than between capital equipment and skilled labour. Consequently, by demanding more skilled labour and being it less substitutable with capital stocks, it is explained how the “skill premium”, namely the skilled labour’s wage over the unskilled labour’s one has increased over time. 2. institutions, that have an impact on inequality through: a. social norms (informal institutions), through the approval of income differences; b. redistributive institutions/policies, like the introduction of a progressive tax system, or with corporate income taxation: for example, lower corporate income tax rates can raise the relative return to capital, which may induce to substitute capital in place of labour, lowering the labour share of income; c. trade agreements, that can affect international trade and migration; d. labour market institutions, that impact on inequality by the institution of minimum wage, unions, pay-setting institutions (pay-per-performance), indexation schemes (“Scala Mobile”, for example, by indexing wages to inflation, caused a sharp decrease of them), Employment Protection Legislation (EPL). 3. interaction between market forces and institutions Lemieux (2008), “The changing nature of wage inequality” Lemieux argued that that wage inequality increases during the 1980’s and the way in which it increased it is incompatible with market forces, hence, it should be addressed to labour market institutions. Indeed, from 1990s (like Krueger (1993)), there was sort of consensus that large part of income inequality comes from SBTC (skill-biased technical change), which is a market force. However, there is also evidence that, keeping advances in technology at pretty similar levels, income inequality raised in some countries and not that much in others. Some scholars suggested that the driving forces of these inequalities are wage-setting institutions. For example, while in Anglo-Saxon countries like US, Britain and Canada, where the wage- setting is at decentralized level and unions are weak, a negative demand change depresses the wages of less- skilled workers, whereas, in France or Germany where wage-setting is centralized and unions are strong, the wages of less-skilled workers remain fixed as well as wage inequality. A second important institution is minimum wage: there is strong evidence that a decrease in the real value of minimum wage played a major role in the increase in inequality in the 1980s. The most relevant institution that had an impact in terms of inequality is de-unionization: the overall effect of unions on wage inequality is usually understood as the interplay between the effect of the inequality enhancing “between” effect (union raise average wages of similar workers) and the inequality reducing “within” effect (union reduce wage dispersion among similar workers). Wage-setting institutions are a fairly successful explanation for recent changes in inequality in the United States. De-unionization implies increasing inequality at the top end but decreasing inequality at the low end, which is consistent with changes in the wage distribution observed over the last 15 years. Adding another institutional factor, the minimum wage, can also account for the fact that inequality also expanded in the low end of the distribution in the 1980s, when the real value of the minimum wage fell sharply. Two other advantages of the institutional explanation are that it is also consistent with long-run trends in inequality in the United States and with cross-country differences in inequality changes over countries. On the historical side, Levy and Temin (2007) provide both historical and quantitative evidence that institutional changes played both a major role in the decline in inequality around and after World War II (the “Detroit Treaty” era) and in the more recent growth in inequality since 1980 (the “Washington Consensus” era). The “Detroit Treaty” era describes the institutional arrangements that took place in the period post-World War II, which determined a period of stable industrial relations in the US economy. This treaty was characterized by high minimum wage, strong support for collective bargaining and unions, high marginal tax rates, Fair Trade Pricing, willingness to regulate industries. From today’s perspective, two features are mostly relevant: a short “guest list”, namely an oligopolistic and regulated structure of industry with concentration of the union power; the fact that business-labour relations would remain a tripartite process with government actively involved as the “third man in the ring”. The Detroit Treaty is more similar to CMEs rather than to LMEs. Given empirical evidence, it is possible to conclude that inequality raised starting from the the period in which Consensus started being applied and one could argue that its policy prescriptions are determinants of such inequalities. Some policy tools can attempt to tackle inequality. 1) Progressivity of tax schedule First of all, according to OECD, there is a strong case for addressing wealth inequality through the tax system. The possible effectiveness of a progressive tax schedule come rom the work of Piketty et al (2014), who provided evidence that, in the US economy: (i) the top 1 percent income share was high before the Great Depression when top tax rates were low (except for a short period from 1917 to 1922), (ii) the top 1 percent income share was consistently low between 1932 to 1980 when the top tax rate was uniformly high, (iii) the top 1 percent income share has increased significantly since 1980 Role of inequality in determining aggregate demand (AD) Modern macroeconomic models are based on the Say’s law, according to which the production of goods creates its own demand, which is another way to say that supply creates its own demand. Consequently, this view suggests that the key to economic growth is not increasing demand, but increasing production. Production (Y, GDP) is given by the sum of profits and wages (namely, income) and, at the equilibrium level, it should be equal to the sum of savings (S) and consumption (C), but, assuming economic agents who optimize over the life cycle (according to the Ramsey model), the amount investments (I) must be equal to the difference between Y and C, which is equal to the level of savings. Consequently, the Aggregate demand function is defined as: Y= S+C=I+C and the supply side (Y=f (L, K, A)) determines GDP. The main implication is that the income distribution, whenever it would be possible to be determined at the individual level, has no impact on GDP, which turns out to be determined by production, namely investments. Moreover, market inequality would provide agents with the right (price) incentives to take sound economic decisions, namely, it allows efficient coordinated actions of the individuals. In line with the Ricardian Equivalence, fiscal policy results ineffective: during any recession, what you can do is to tax less in the current period, but taxing at a higher level in the following period, without any relevant change in consumption and, therefore, in GDP. This is because agents are assumed to be rational and, with their rational expectation, they do expect higher taxation in the following period, leaving their level of consumption unchanged. A Keynesian policy maker, on the other hand, would suggest to increase public expenditure with a decrease in taxation, generating deficit, stimulating consumption and the aggregate demand. Indeed, the Keynesian model does not assume that the level of savings (S) is equal to the level of investments (I): this is due to the presence of animal spirits in the market, namely waves of optimism or pessimism among individuals, that result at an aggregate level (different from proper rational expectations). There is a vicious circle: a decrease in AD (interpreted as the determinant of a recession period) causes a decrease in GDP, which causes a decrease in the level of investments and consumption, which further lowers the aggregate demand level. There follows that, in the case of low AD, the productive capacity of an economy may be not fully used. In other terms, whereas the “Supply side economics” (Say’s rule) claims that GDP (namely efficiency or production) has an impact on AD, Keynesian economics claims that is the even AD that determines GDP. Anyway, especially during recessions, it is crucial to understand what are the determinants of the Aggregate Demand. According to the Keynes model, the level of aggregate demand has an impact on GDP and on employment and inequality is a key determinant of the aggregate demand, because highest income earners typically spend a lower share of their income in consumption, rather than other earners’ groups would do. As a consequence, there are unclear theoretical conclusions on the desirable institutional framework due to the disconnection of equity and efficiency that can lead to wrong policy implications. According to the classical theory of employment, the equilibrium levels of employment and unemployment are determined in the labour market from the intersection of labour supply (that derives from the maximization problem of an individual that has to choose how much to consume, given that his income derives from work and the optimal quality of work depends on the trade-off between leisure and working hours, according to the wages that he will earn: in other terms, the marginal returns from work) and labour demand (the first derivative of the production function with respect to labour or the marginal productivity of labour), given a certain level of real wages (possibly affected by the frictions in the market). An increase in the real wage (w-p) causes an increase in unemployment. As a consequence, in the classical model, in order to understand the determinants of unemployment, it is necessary to study the specific labour market and its frictions, which define the level of real wages. According to the Keynesian model, instead, the aggregate demand determines production and, consequently, also the level of employment and, given firms’ labour demand, aggregate demand also determines real wages. The “Wage Schedule” curve represents the amount of money that is necessary in order to fire workers to invest the same amount in other productivity factors and, to some extent, coincides with the labour demand since they will not pay them more than their marginal productivity. Consequently, real wage and labour supply determine unemployment, which is a result of not only labour market forces. Consequently, the remedy in the case of unemployment, in the Keynesian model, is to provoke an expansion in the Aggregate Demand: if AD increases, the labour demand increases, the level of real wages will be lower and the employment will increase. Moreover, if real wage is lowered without an increase in AD, there is no real effect real variables and possible deflation. Despite the models above suggest differently, Krugman, with an article in the New York Times (2015) argued that the market of labour cannot be compared to the one of butter and, hence, it is not necessarily true that an increase in minimum wages will lead to higher unemployment levels (and there is empirical evidence of this). Indeed, by raising minimum wages by a substantial amount, it would be possible to reduce within-firm turnover, making them more productive, direct monetary and fiscal policy towards full employment. The argument is that not only inequality, but also taxation, public expenditure and other variables do affect Aggregate Demand. Moreover, wages are not simply determined by the marginal productivity of workers (that would be the case if the labour market was without any friction) but, instead, by the social interaction in the market. It is implied that institutions are the key determinants of AD both directly, through public deficit, and indirectly, by shaping personal and functional income distribution. In general, institutions are very important due to their impact in short-term economic outcomes. Consensus institutions might have detrimental effects on macroeconomic outcomes, in the case in which the hypotheses of the Keynesian model are correct, as those institutions let inequality increase among individuals, which can lower GDP through lower Aggregate Demand. In the presence of a Keynesian recession (a decrease in AD) policy prescriptions should be antithetic to those delivered by the Consensus: if Keynesian hypotheses hold and there is also a causal effect starting from the Aggregate Demand to the GDP, we should have policies on public expenditures and on public debt that cause an increase in consumption and, then, in the AD (which is the case if the assumption S=I does not hold); redistribution through, for example, the European Welfare State, which pushes consumption up since lower-income people will consume more than highest-income earners; labour market institutions, that have an impact on inequality by increasing the minimum wage and the bargaining power of workers in the labour market, or by defending trading unions. The debate about inequality historically focused on the case of United States, where it was mostly evident the move from the Detroit Consensus to the Washington Consensus, together with the impoverishment (or even disappearance) of the middle class: this was cause by the fact that inequality caused lower disposable income for large number of people. Indeed, whenever resources are switched from the largest portion of population to the top 10%, this top 10% will not invest in consumption but rather on derivatives or other financial products. Thus, social consumption decreases, AD decreases and GDP as well. In the US, this switch was kind of dramatic with not comparable situations in other countries. However, there are other two possible consequences of inequality. Firstly, the lower income for sub-classes in society might be counterbalance with the attempt of sustaining consumption by the promotion of private debt. This actually happened in the US, with an increase in the consumption of goods but also of financial products. Alternatively, in order to counterbalance the decrease in aggregate demand, one national economy could adopt an export oriented growth’s strategy, in the idea that trying to allocate national production in abroad markets needs an adequate level of efficiency and competitiveness. The solutions of promoting private debt and export oriented growth can be actually effective and, hence, they are very commonly adopted. In this picture, it is shown that personal consumption expenditures are actually the major determinant of the gross domestic product at an increasing rate over time. This is evidence of the fact that personal consumption became a more and more important component of the total GDP in the same period in which, in the US, inequality started increasing, namely, in the same period in which Washington Consensus ideas were more shared.