Jorge Braga de Macedo

OECD and FEUNL

1. Introduction

The sources of, and policies for, Portuguese economic development were discussed at a Bank of Portugal conference held at the Gulbenkian Foundation in Lisbon on May 24 and 25, 2002 - where an earlier version of this text was presented. Such long term issues may appear removed from the short term constraints facing Portuguese taxpayers, yet they are at the heart of the determinism called fado at that conference (Miguel Freitas, 2002). Rather than looking at policies, citizens often fall into forms of historical and geographical determinism which have little to do with what actually happened.

This is why I want to bring some work done at the OECD to bear on the issue of the Bank of Portugal conference, drawing on work prepared for the forthcoming 40^th anniversary volume of the OECD Development Centre, titled Development Redux. I will do so from a time and then a space perspective, hoping that through the combination of both we can derive some policy lessons, and especially we can address some of the trade-offs involved. That is, after all, what economists should help with.

First I review briefly the Portuguese convergence record, combining my investigation of the inheritance of the real with the global perspective of Angus Maddison (2001). Then I establish the role of expectations in the balancing act between market and government failures under imperfect information. It is widely acknowledged that this balancing act cannot neglect history but - because it involves institutional change - is also determined by expectations. Even when there is a decisive element of self-fulfilling prophecy, the role of expectations differs from central planning to the extent that it is consistent with information available in world financial markets and cannot therefore be manipulated by national authorities. In other words, the fulfilment of the prophecy is credible. Thus hope, translated into economic analysis, implies an "overlap" between history and expectations as potential determinants of the development path. This basic pitfall in historical determinism is underlined in section 3, following my (1990) and (2002).

Ongoing work by Daniel Cohen and Marcelo Soto (Cohen, 2002), on the relative contribution of human capital, physical capital and total factor productivity in explaining the income per capita gap, using new series on human capital that account for the quality of education and for health is summarised in section 4. As the three channels are equally relevant, there is certainly no geographic determinism either. In the concluding section 5, I introduce institutional change as supporting, or not, policy credibility in development. Neither historical inheritances nor geographical yardsticks can neglect expectations, therefore confirming the pitfalls in determinism, Portuguese or from anywhere else.

Portugal has revealed an historical preference for hidden taxation, including the one achieved through inflation and excess borrowing by the state. Thus domain revenues associated with the overseas trade monopolies and collected mostly in Lisbon were more significant than those associated with income taxes, levied since 1641 (Macedo et al 2001a). Centralised revenues were easier to administer and they helped maintain monetary stability in the face of recurrent military expenditures. However, growing domain revenues gradually narrowed the domestic tax base and divorced taxpayers from transfer recipients - features which came to characterise the fiscal constitution of the 1700s. Uunfortunately they remained even after the combination of foreign and then civil wars, monetary instability and collapsing revenues stunted Portugal’s economic development. The 1820 liberal revolution was associated with the loss of social confidence and financial reputation, making the taxation of capital and the redistribution among social groups difficult. Civil rights were seen as inimical to monetary stability, so that political and financial freedom clashed until the gold standard brought nominal and real convergence with what is now the European average from 1854 until the 1891 financial crisis. Exit from gold standard initiated a period of maximum divergence, which lasted until the postwar European integration boom, which coincided with unprecedented growth in the world economy as a whole.

The annual estimates of population, gross domestic product and GDP per capita for 124 countries, 7 regions and the world from 1950 to 1998 in Maddison (2001) provide a yardstick – albeit imperfect. According to the purchasing power standard used (1990 international dollars), world GDP per capita grew continuously during the period except for 1975, 1982 and 1991, when it dropped by 0.4% in each one of these years. In Europe (defined as EU+EFTA) the years of negative growth were 1975 (-1.0%), 1981 (-0.1%) and 1993 (-0.8%).

Portugal’s GDP per capita relative to the European average declined during the three world recessions, typically with a lag of one to two years. The average rate of decline was 4.5% in 1974/75, 3.3% in 1983/84 and 0.5% in 1993/94. Two episodes of sustained convergence can be identified, from 1959 to 1973 (22 points of GDP in 13 years) and from 1984 to 1998 (14 points in 12 years). During the 1950s and between the two convergence episodes, Portugal exhibited a slower than average growth in standard of living.

It turns out that neither the redistributive revolutions of 1910 and 1974 nor the ensuing political instability changed the fiscal constitution much. From 1960 until 1985, EFTA allowed Portugal to develop an export base in manufacturing (sometimes called a "pyjama republic"). But European integration did not balance mutual political responsiveness with economic interdependence. Public and private interest groups seeking transfers from the state and thereby holding on to the tax base took advantage of European structural funds. The whole process reinforced some of the interests vested in state intervention.

Indeed, economic and political integration have generally not been balanced, except when meeting the convergence criteria for the single currency became an explicit policy objective, shortly before the first EU Presidency in early 1992. The synchronicity brought to the fore the view of Portugal as a "good student" of European integration. While distancing itself from that view, the government elected in late 1995 made it a political imperative to join the eurosystem in 1998. Shortly thereafter, the imbalance between economic and political integration returned. The reason for this imbalance varied, but, until 1989, the public sector was frozen by a constitutional ban on widespread privatisation. This reinforced some of the worst features of the fiscal constitution, namely a traditional preference for hidden taxation, including through inflation and a trend towards increasing public expenditures which has remained unchecked in spite of membership in the euro.

The role of expectations is recognised in the responses of two Nobel laureates to the letter dated 15 February 2000, where the managing editors of the European Journal of the History of Economic Thought asked them to list the five most significant works on economics in the twentieth century (Editorial 2001, p.285). The work of Robert Lucas, whose 1985 Marshall lecture (1988) revived interest in growth theory and launched a large literature on "endogenous growth", is relevant here. Yet, many development economists might start evoking their field with history rather than with expectations. Indeed, history alone determines the equilibrium under three different environments. First, "if the future is heavily discounted, individuals will not care much about future actions of other individuals, and this will eliminate the possibility of self-fulfilling prophecies." In other words, the more impatient individuals are, the more they are prisoners of history! Second, "if external economies are small there will not be enough interdependence among decisions". Third, if "the economy adjusts slowly, then history is always decisive. The logic here is that if adjustment is slow, factor rewards will be near current levels for a long time whatever the expectations, so that factor reallocation always follows current returns" (Krugman 1991, p. 664).

Suppose that labour in the increasing returns sector (X) is less than in equilibrium then the shadow price placed on the "asset" of having a unit of labour in this sector rather than in the constant returns sector (C) would have to rise in a deterministic path. Suppose it does not because everyone expects the wage to rise in future at a rate (q) just enough to compensate for its lower value at present. The dynamics are then given by an oscillation towards the two equilibria with complete specialisation in X or in C (the system has two complex roots with positive real parts as described in Figure III in Krugman 1991; see a diagram in the same vein in my 1990).

In an application to urban development in the United States, Timothy Harris and Yannis Ionnides (2000) interpret the "asset" of having a unit of labour in the increasing returns sector rather than in the constant returns sector as farmland values and housing values. They then test whether those asset prices anticipate, or instead follow, urban development. The result across over 150 cities is that history "dominates the process by which one city becomes a metropolis and another languishes in the periphery". Nevertheless, there remains the possibility of asset prices anticipating population as expectations of future development are incorporated in asset prices for a particular city. The empirical verification that farmland values and housing values did not anticipate urban development reinforces the importance of institutional change (see section 5 below).

As expectations include the tendency towards convergence, they impose tighter and tighter constraints on inadequate policies. Also, even though future generations are not represented in majority voting, greater awareness of the need to implement sustainable policies brings pressure on elected governments to clarify the intergenerational effects of current policies. This applies to the physical and cultural environment, as well as to the provision of public goods and transfers through taxation. The awareness is also rising that excessive taxation, whether overt or hidden in the form of inflation, discourages saving and stifles growth. This may appear not to be a developing country problem, but the difference arises mainly in the mix between overt and hidden taxes, as the latter dominate in developing countries.

As growth prospects fall due to the absence of incentives to save and invest, so does employment, reducing future consumption and increasing social deprivation. In due course these policies will be corrected. Yet, without adequate institutions, there may be reversions into inadequate policies.

In that sense, economic adjustment helps prevent policy reversals for any given level of interdependence in time (low discount rate) and in space (large externalities). Conversely, high interdependence (coming from more patience or strong externalities) may overcome given costs of institutional change and adaptation.

Using new data on human capital, this section shows that there is not a unique factor behind the poverty of nations. Poor countries are "slightly" disadvantaged in each one of the fctors behind prosperity. But the combination of these slight weaknesses results in huge income gaps, as shown by Cohen (2002). A standard neoclassical production function, augmented by human capital, may be written as:

y = A(k/h)^ah (2)

with y=Y/L and k=K/L. Equation (2) suggests explaining income differences between rich and poor countries by three different terms: differences in human capital stocks (h), differences in physical capital stocks (k/h)^a, and differences in productivity (A). This decomposition results in the following table (figures are relative to rich counties' average):

Source: (1) through (4) Cohen and Soto; (5) Hall and Jones (1999)

Table 1 shows that, excluding Sub-Saharan Africa, average income per capita in poor countries is only 25% that of rich countries (the capital share is assumed equal to 1/3, which is a standard assumption in growth accounting). What is behind this difference? Columns 2 to 4 show that there is no single reason explaining this income gap. Human capital is 58% that of rich countries, while the relative shortage of physical capital is of 65%. Finally, total factor productivity is just 65% lower than rich countries'. Put simply, we can say that poor countries (excluding Sub-Saharan Africa) are, on average, a third poorer than rich countries in each one of the three terms forming wealth. Although the gap in each one of these terms individually does not seem disproportionate, their combination results in an income gap of 75%.

The case of Sub-Saharan Africa is even more spectacular. This group of countries has only 40% of rich countries' level of each, human capital, physical capital and productivity. This scarcity implies that average income is just 6% that of the rich world.

Column 5 shows total factor productivity averages inferred from Bob Hall and Charles Jones (1999). The figures that they report for A are not directly comparable since they assume a production function given by:

Since a=1/3, their values for A are raised to the power 2/3 to make them comparable to column 4. The differences between both measures of total factor productivity are minor.

Calls for an interdisciplinary approach to development and the emphasis on its internationally agreed goals (as recorded lastly in the Monterrey declaration on Financing for Development) should not obscure the essential prerequisite of higher economic growth. In spite of agreement that market-based economic growth is key for the prevention of poverty and hunger, discussion continues about which kind of economic growth strategy to follow in developing countries. Sometimes this discussion tends to be fairly superficial, especially in a developing country context, by solely focusing on macro-economic conditions and the functioning of markets in a narrow sense. What is needed is a focus on the broader institutions of a well-functioning market economy: legal, political, social and cultural.

There are many specific examples that governance and institutions matter for development (Justin Lin and Jeffrey Nogent 1995; José Tavares 2002 has a series of remarkable comparisons between Portugal, other cohesion countries and the Asian tigers with respect to firms, financial markets and laws).

Nevertheless, little is known about how the independence of the central bank and appropriate budgetary procedures interact with political accountability in particular institutional settings is not known. Torsten Persson, Gerard Roland and Guido Tabellini (1997) do find a general trade-off between independence and accountability which provides support to the separation of powers argument from eighteenth century political philosophy. In particular, the separation between executive and legislative powers is applied to the budget process as an illustration of the benefits of democratic governance.

Building on their notion of complex interdependence, Robert Keohane and Joe Nye (2000) show that, with the spread of free information, the credibility of policy becomes essential - a direct consequence of the role of expectations stressed in section 1 above. Nevertheless, there are few applications of these insights to developing countries, so that the burden of the initial conditions makes institutional change less credible.

Re-reading what I wrote in my (1990), "Portugal may have moved from eurotimidity to lusoconfidence, even though a reversal cannot be excluded", I cannot help seeing such reversal in the policies that brought about the current budgetary squeeze because they greatly reduced the speed of adjustment and of institutional change for fear of vested interests (Daniel Traça, 2002 provides a specific example coming from labour market institutions, where high flexibility and low unemployment is observed together with low adaptability and productivity).

This fight for the common good is the typical development challenge, and Portugal is facing it again today. Looking at what happened ten (or indeed two hundred) years ago, comparing to other countries in the EU and elsewhere is important. But only hope in future development avoids the pitfalls in historical and geographical determinism.

REFERENCES

Cohen, Daniel (2002), "Growth in Theory and Practice", Development Redux, OECD Development Centre, forthcoming

Editorial (2001), Symposium on the Most Significant Contributions to Economics during the Twentieth Century: Some Nobel Laureates' Views, European Journal of the History of Economic Thought, Vol 8 (3) Autumn, p. 285

Freitas, Miguel Lebre de "Crescimento Económico Português: Um Manifesto Anti-Fado", paper presented at a Bank of Portugal conference on Portuguese Economic Development in the European Space: Sources and Policies held at the Gulbenkian Foundation in Lisbon on May 24 and 25.

Hall, Robert and Charles Jones "Why do some countries produce so much more output per worker than others?", Quarterly Journal of Economics, 114 (1), pp. 83-

Harris, Timothy and Yannis Ionnides (2000), History vs Expectations an empirical investigation, Tufts University Department of Economics Discussion Paper 2000-14

Keohane Robert and Joseph Nye (1999), "Power and Interdependence in the Information Age" in Elaine Ciulla Kamarck and Joseph Nye, editors, democarcy.com? Governanvce in a Networked World, Hollis Publishing Company, NH

Krugman, Paul (1981), Trade, Accumulation and Uneven Development, Journal of Development Economics VIII 149-61

Krugman, Paul (1991), History vs Expectations Quarterly Jouranl of Economics n° 2, May, pp. 651-667

Lin, Justin Yifu and Jeffrey Nugent, "Institutions and Economic Development" (1995), in Jere Behrman and T.N. Srinivasan, editors, Handbook of Development Economics, volume 3A, North Holland, pp. 2301-2370

Lucas, Robert (1988), "The Mechanics of Economic Development", Journal of Monetary Economics

Macedo, Jorge Braga de (1990), "Preface" to João Cesar das Neves, Da validade científico do conceito de equilíbrio de pobreza, Lisbon: Ministry of Finance, pp. 11-14

Macedo, Jorge Braga de (1999), "Portugal's European Integration: the limits of external pressure", Nova Economics Working Paper nº 369, December.

Macedo, Jorge Braga de (2002), "Comparative Development and Institutional Change", OECD Development Centre, March

Macedo, Jorge Braga de, Álvaro Ferreira da Silva and Rita Martins de Sousa, War, taxes and gold: the inheritance of the real, in Transferring Wealth and Power from the Old to the New World, edited by Michael Bordo and Roberto Cortes-Conde, Cambridge: Cambridge University Press, 2001a.

Maddison, Angus (2001), The world economy: a millenial perspective, Paris. OECD Development Centre.

Persson, Torsten, Gérard Roland and Guido Tabellini (1997), "Separation of Powers and Political Accountability", The Quarterly Journal of Economics, Vol. CXII, for Harvard University by the MIT Press, Cambridge, Massachusetts

Rosenstein-Rodan, Paul (1943), "Industrialisation of Eastern and South Eastern Europe", Economic Journal, vol. 53

Tavares, José (2002),"Firms, Financial Markets and the Law ", paper presented at a Bank of Portugal conference on Portuguese Economic Development in the European Space: Sources and Policies held at the Gulbenkian Foundation in Lisbon on May 24 and 25.

Traça, Daniel (2002), "Labor Markets in Portugal: Recent Performances and Challenges for Development in the European Context", paper presented at a Bank of Portugal conference on Portuguese Economic Development in the European Space: Sources and Policies held at the Gulbenkian Foundation in Lisbon on May 24 and 25.