Weather

Saturday 16 January 2010

Energy and Democracy: an Economic Perspective

The problem is that the good Lord didn’t see fit to put oil and gas reserves where there are democratic governments.
Dick Cheney, Interview to the "Washington Post", 30th of July, 2000.

17 October 2007, by Paolo Figini | 1 document

Introduction

The relationship between energy reserves and democracy, or in more general terms, between natural resources and political institutions is one of the most debated topics in social sciences. Such issue lies at the junction of two communities of researchers which often use different assumptions, interpretative models and methodologies: the economists and the political scientists. However, neither economists nor political scientists would agree with the US vice-president citation reported above. The supposed incompatibility between energy resources and democratic institutions, if supported by data, can not stem from divine obscure will, nor be a pure coincidence. There must be political and economic reasons.

In this brief survey, we want to summarise the main theoretical and empirical evidence stemming from the literature [1], highlighting the limits and discussing a few paths that could be taken to improve the robustness of the findings. In the next section we will briefly review the so-called natural resource curse argument, with its supporting evidence and the models suggested to explain it. Since one of the main channels through which the resource curse is supposed to work is the quality of institutions, in section three we will finally focus on the relationship between energy resources and democracy.

The Natural Resource Curse

Sachs and Warner (1995 and 2001) inquire into the relationship between a country’s natural resources and its economic performance. Counter to the intuition, in post world-war II, they find evidence of a negative relationship: resource abundant countries have grown on average less than resource poor countries. Resources, they argue, are not a blessing which helps economic development, but rather a curse, limiting growth and socio-economic progress. Since their 1995 seminal paper, those results have been discussing by a flourishing literature. According to Cabrales and Hauk (2007), the empirical evidence, somehow weak and with many caveats, can be summarised in these five stylised facts:

  • Fact 1 – Countries rich in natural resources grow slower on average than natural resource poor countries (the natural resource curse argument).
    However there are some resource rich countries that have grown very fast and for which natural resources are indeed a blessing: this leads to:
  • Fact 2 – The evidence is inconsistent with a monotonic effect of resources on development/growth (see Robinson et al. 2006).
    The theoretical and empirical analysis has since been focussing on the reasons why natural resources are somewhere a blessing and somewhere else a curse. Policy failures have been identified as the main cause:
  • Fact 3 – The quality of institutions is a decisive factor in determining whether natural resources are a blessing or a curse (see Mehlum, Moene and Torvik, 2006).
    Empirical findings also suggest a reverse causality:
  • Fact 4 – Natural resources have antidemocratic properties: oil and mineral wealth tends to make countries less democratic (see Ross, 2001) and
  • Fact 5 – Many revolutions are linked to rents derived from natural resources (see Collier and Hoeffler, 1998 and 2005).

What explains the curse? As Sachs and Warner (2001, p. 833) clearly state “[m]ost current explanations for the curse have a crowding out logic. Natural resources crowd-out activity x. Activity x drives growth. Therefore natural resources harm growth”. Gylfason (2001) lists four different channels of transmission “x” that have been used to explain the way in which resource abundance leads to the under-performance of the economy. First, a natural resource boom (i.e., the discovery of a new oil-field), with its strong effect on exports leads to an over-evaluation of the national currency. At the same time, the great inflow of income into the country boosts domestic demand and starts inflationary pressures, particularly on non-traded goods. Both phenomena crowd out traded sectors, in which profits shrink because they sell at international given prices, thus hurting the overall development of the country. Such a negative stagflationary spiral is the so-called Dutch disease.

Second, countries which rely on their natural resources may, inadvertently or deliberately, neglect to invest in human resources. Those are, in contemporary economies, the most important fuel of sustained growth, thus natural resource rich countries being caught in a poverty trap. Third, as a generalisation of the previous rationale, natural resource abundance may give citizens and governments a false sense of security, leading to the non-implementation of efficient and growth-enhancing economic policies [2]. As Gylfason (2001, p. 85) puts it “rich parents sometimes spoil their kids. Mother Nature is no exception”.

These three theories highlight the lack or the ineffectiveness of economic policy. However, there is a fourth explanation which considers the interplay between natural resources and the “quality of institutions”. According to Robinson et al. (2006) and to Mehlum et al. (2006) the impact of resources on growth crucially depends on the quality of the political institutions and, in particular, on the degree of corruption in the public sector. Mehlum et al. show that the variance of economic growth among resource rich countries is due to how rents are distributed. Some countries have institutions that favour producers and re-investment of those rents, while others have “grabber friendly” institutions that divert scarce entrepreneurial and human resources into unproductive rent-seeking. Therefore, the quality of institutions is the factor which determines whether abundance in natural resources is a curse or a blessing for a country.

However, in this fourth model an endogeneity problem appears, and the reverse causality has also to be considered: resource wealth can lead governments and producers to implement rent-seeking activities, thereby leading to corruption and “grabber friendly institutions”. In extreme cases, to avoid the redistribution of generated wealth, rent-seeking can take the form of authoritarian governments. This causal link from resources to democracy, which could be called the natural resource political curse, will be tackled in the next section.

As the theory, also the empirical analysis joins the results of two different strands of research. First, the huge literature on endogenous growth of the last twenty years has been identifying the main engines of economic growth. It is well known that the level of human capital (and therefore the extent of investment in education) and the quality of institutions are two of the most important variables at play (on this last point see Persson and Tabellini, 2006, for a survey). There is no reason, hence, to doubt about the “’x’-to-growth” link described above.

The second, more specific, strand of literature attempts to assess the validity of the link between natural resources and the “x” proposed channel of transmission. Among others, Stijns (2003), Hausmann and Rigobon (2003), and Sachs and Warner (2001) find (weak) evidence for the Dutch disease effect [3]. Gylfason (2001) shows that natural capital crowds out human capital, thereby slowing down the pace of economic development. However, Bravo-Ortega and De Gregorio (2006) find the link not to be linear: the negative effect of resources on growth can be offset by high education levels, thereby making natural resource abundance a blessing for countries with high human capital. This last result can explain the performance of countries like Norway.

Recently, more emphasis has been put on the “quality of institutions” effect. The seminal paper by Sachs and Warner (1995) considers the specific mechanism which leads resource abundance to a deterioration of institutional quality, but dismisses it as empirically unimportant. On the contrary, Torvik (2002) finds evidence that natural resources increase rent-seeking activities, thereby lowering growth. This channel of transmission introduces to a more political field of research which relates to the (suggested) causal effect of natural resources abundance on democracy. Is there a natural resources political curse? This is the question to be tackled in the next section.

Does Oil Hinder Democracy?

Does Oil Hinder Democracy? This is the title of an influential paper by Michael Ross (2001). Ross perspective differs from the literature analysed in the previous section for at least two reasons. First, being a political scientist, not an economist, Ross perspective is somehow different: he does not look at the interplay between resources and institutions in determining the economic performance of a country, but focuses on the causal relationship between natural (primarily energy) resources and democracy. He then attempts to find evidence supporting the different theories brought forward to explain this link. According to Ross and coherently with what stated in the previous section, such theories can be sorted into two categories: those suggesting that oil-generated wealth causes governments to do a poorer job in promoting economic development (through ineffective education policies, corruption, etc.) and those suggesting that oil-generated wealth makes countries less democratic; with Ross, our survey now follows this second line of research.

Second, he unfolds the analysis around two different dimensions: the geographic dimension, attempting to test whether any causal link is due to the (non) democratic performance of particular groups of countries with some cultural and historical characteristics (i.e., the Middle East and its own Islamic culture); the sectoral dimension, with the estimate of the effect generated by different types of primary resources.

Ross introduces the three main causal mechanisms that can explain the link between energy rich countries and authoritarian rule: the rentier effect, the repression effect, the modernization effect, which now we briefly explain.

  • The rentier effect. Countries rich of gas, oil, or other natural resources gain much of their revenues from stamps, royalties, or export taxes. Such rent-seeking activities can be used to relieve social pressure that might otherwise lead to demand of greater accountability. This may occur in three ways. The first is through a taxation effect. Since governments get sufficient revenues from oil or other resources, they do not have to impose tight budgets and / or heavy tax burdens on citizens, thus relaxing their control over government activity and making fiscal pacification much more effective. Second, through a spending effect, oil revenues can be (at least partially) used for general spending and for patronage, thus dampening latent pressures for democratization. Third, through a group formation effect: governments can use their money to corrupt or prevent the formation of independent social groups that may be inclined to demand political rights.
  • The repression effect. This (much more political) story unfolds as follows: oil-generated wealth may induce governments to spend more on internal security, halting the population’s democratic aspirations. Since the government has more money available, it is less costly to be authoritarian in order to avoid the sharing of its rent with the population. This effect is consistent with Collier and Hoeffler (1998) findings that natural resource abundance tends to make civil war more likely, since the expected benefit of winning the war is much higher for the different stakeholders at play (governments, oppositions, political parties, ethnic groups, oligarchies).
  • The modernization effect. Democracy stems from a collection of social and cultural changes, including occupational specialisation, urbanisation and high levels of education. None of these changes are likely to occur in a resource abundant country: the labour force is trapped in the energy sector and does not shift to the manufacturing and to the service sector, citizens are not forced to move to the cities and the lack of economic need diminishes the effort of investing in human capital. As a side effect, it is known than democratic pressure is low when there are less educated citizens, thereby reinforcing the effect.

Ross empirical findings are consistent with the theory and can be summoned as follows: first, oil and other mineral fuels (gas, coal) do hurt democracy. Second, energy resources inhibit democracy in poor countries while in rich countries such link is not statistically significant. Third, the link is not specific to the Middle East but valid at the global level. Fourth, all resources, included non-fuel mineral wealth, hinder democracy. Fifth, there is weak evidence to support all three channels presented above, the rentier, the repression and the modernisation effect.

Discussion

To conclude this brief survey, does not only resource wealth reduce economic growth, but also curses through hindering democratic institutions: thus, the trap of natural resources has both an economic and a political dimension.

Still, results provided by this strand of literature are not particularly robust and many roads remain to be explored. With respect to theory, the high number of variables at play and their complex interactions call for deeper analysis. Cabrales and Hauk paper (2007) is a firm move in this direction. Empirically, there are many caveats, both methodological and data related, as Brunnschweiler (2007) argues.

As regards data, it has been claimed that natural resources are not properly measured: most of the papers use the export share of primary exports in GDP as an index of resource abundance. There are several limitations in this measure: first, a measurement error appears, since GDP is modified by a change in the export performance; second, exports measure the flow of resources outside the country. They do not measure neither the stock of available resources (the abundance) nor the present value of such stock (the wealth). Empirical analysis should take care of this discrepancy with the theory; third, many papers identify natural resources with the whole primary sector, others select the mineral sector, others the mineral-fuel sector. Although it is claimed that the results are robust to the selection of any of these measures (Sachs and Warner, 2001), the literature should focus more on the energy sector, since most of the theoretical work is concerned with fuel-related rent-seeking or, more in general, with abundance of those minerals that are important inputs in contemporary production, such as copper and silicon [4].

A second critique is about the econometric methodology used: most of the works on the natural resource curse use cross-country regressions. However, since theoretical models describe mechanisms involving dynamic effects within the country, panel techniques (i.e., fixed effects, random effects, the Arellano-Bond estimator) should be used. The dynamic process is particularly important in the story relating energy and democracy, for which the discovery of natural resources should lead, with a lag, to the detriment of democracy and to the worsening of the quality of institutions. Ross (2001) goes in this direction by using pooled cross-section time-series data; however, more could be done on this aspect.

A final word should be said on democracy, such a blunt concept to be measured. Although this is a general critique to all the empirical research on the quality of institutions, the many indicators used for proxying democracy, corruption or the quality of institutions have many caveats, and do not address properly the multi-dimensional characteristics of such concepts.

These limitations are also the strength of this research field: the link between energy resources, growth and democracy has still so many aspects, peculiarities, and perhaps surprises that have to be discovered and correctly tackled by the scholars.


P.S.

References


http://www.cartografareilpresente.org/article133.html
Posted by Ptahhotep at 21:30

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)