Given these developments in growth theory, a whole new direction of research opened up in the 1990s. Pritchett (2006) describes how researchers – particularly in the World Bank – began to run numerous cross-country Figure 2.1. Regional GDP growth rates, 1960–2006.
regressions with growth as the dependent variable (on the left-hand side of the equation) and endogenous factors on the right hand side of the equation, as the explanatory variables.
As we shall see, this work has produced some clear results. However, Pritchett (op. cit.) urges caution in the interpretation of these results. First, regressions of this form may be able to establish the existence of a negative relationship between growth and, say, corruption, but specific policies do not flow directly from this. Second, as with all cross-country studies of this type, the idiosyncratic characteristics of each country may, to some extent, cancel each other out, so that the aggregated results do not hold at the country level for many individual countries. Third, the regressions may establish a long- term relationship – even a causal one – but this says nothing about where the ‘turning points’ in the process are, or how they can be triggered. That is, sound institutions are positively correlated with higher growth rates over a long period (up to thirty years), but this does not tell policy-makers what specific reforms could be implemented in the short term to lead to an acceleration of growth.
Fourth, and perhaps most importantly, the results of this work have largely failed to explain the divergent growth experience of the developing regions in the 1990s. On the negative side, the decade was characterised by a series of severe financial crises, which had a devastating impact upon growth rates.
These could neither be predicted nor explained by this body of work. On the positive side, a number of countries – notably China, India and Vietnam – performed very well in terms of growth, but again this could not be explained by the results of the regressions that have been described.
Therefore, as has been stressed throughout this chapter, it is clear that cross-country regressions of the kind detailed below can offer some insight into those endogenous features of the economy that may encourage (or dis- courage) growth. However, there will be countries that have all the ‘right elements’ in this sense, yet do not see accelerating growth, and others where seemingly few of the ‘necessary’ preconditions for growth are in place, yet growth does accelerate. As ever, policies cannot be implemented on one-size- fits-all basis, but need to be tailored for the different circumstances of each country.
With these caveats in mind, the final section of this chapter considers what we can say about the relationship between the elements of developing coun- tries’ financial systems and the rate of economic growth of these economies.
2.4.1 Finance and growth
Although this section focuses on the relationship between financial sector development (FSD) and growth, it is worth noting that, for many observers, the key variable in determining the divergent rates of growth that developing countries have seen is the quality of their institutions. The conclusion that
‘institutions rule’ was reached by Rodrik, Subramanian and Trebbi (2004),
who demonstrated that the primary long-term determinant of a country’s level of income was the quality of its institutions. Clearly, it is difficult to judge institutional quality in this sense, and the term ‘institution’ can also be defined either narrowly or broadly, but for the authors, it is these internal features of an economy – rather than their geographical location or integration in the world economy, for example – that determine long-term growth rates.
In addition to the political, legal and administrative framework, a coun- try’s financial infrastructure and institutions have also been associated with long-term growth. Although the idea that the financial system is a key driver of growth has, as we have seen, only come to receive attention relatively recently, it is by no means a new concept.
At the beginning of the twentieth century, Joseph Schumpeter11 stressed the important role played by the banking system in allocating savings towards the most productive use, and thereby improving productivity and thus eco- nomic growth rates.
Box 2.6 Joseph Schumpeter
Joseph Alois Schumpeter (1883–1950) was an Austrian-born economist and political scientist, who is widely regarded as being one of the most influential political economists of the twentieth century. Schumpeter coined the phrase ‘creative destruction’ to describe the processes of capitalist development and was particularly influential with regard to the theory of business cycles. Schumpeter was critical of the approach taken by mainstream neo-classical economists, seeing it as too abstracted and idealised to be capable of capturing reality. Indeed, Schumpeter pointed out that ‘equilibrium’ in the standard ‘Walrasian’ sense is inherently static and so has nothing to say about economic growth and development. For Schumpeter, it is the entrepreneur who drives eco- nomic progress: although many will fail, some will succeed (i.e. creative destruction, where old methods and practices are continually replaced by new methods). Thus capitalist growth is maintained. However, Schumpeter also argued – like Marx – that capitalism contained the seeds of its own demise. Ultimately, the development of an educated middle class and an intellectual elite would lead – in a democracy – to the implementation of policies inimical to the entrepreneur. Though not a socialist on principled grounds, Schumpeter predicted that some form of socialism was therefore the destination of modern capitalism.
In the 1950s, Arthur Lewis12 argued that the relationship between econo- mic growth and financial development ran in both directions. That is, an
increasingly sophisticated financial sector developed as the economy grew (growth caused financial sector development), and itself acted as a spur to further economic growth (financial sector development caused economic growth). This causality question remains a key issue, as we shall see below.
In the late 1960s Raymond Goldsmith13 also emphasised the connection between the real and financial sectors, arguing that financial markets encour- age growth through efficiently allocating resources through time.
What has more recent research to say on the process through which FSD may spur growth, however? In this regard, two channels are generally identified:
1 through its impact on capital accumulation, both human and physical;
2 through its impact on the rate of technological progress.
Both of these channels may facilitate growth through the intermediation of savings and their allocation to productive use within the economy, through investment in physical capital, education or training on the one hand, or through research and development on the other.
The key actors in this process are third-party financial intermediaries.
Levine (1997) identifies five key functions of these institutions that enable them to perform this role:
1 the mobilisation of savings;
2 facilitating the management of risk;
3 obtaining and filtering information on investment opportunities. (i.e. ex ante monitoring);
4 monitoring the use of funds (i.e. ex post monitoring);
5 facilitating the exchange of goods and services.
As well as these features, financial sector development may also be a spur to economic growth through its impact on transaction costs, which increases the economic viability of many productive ventures. That is, the higher the trans- action costs involved in a commercial venture then the higher the rate of return needed to make the venture viable. As transaction costs fall, there- fore, ventures that were previously not viable, in this sense, increasingly become so.
Considerable empirical research has shown a strong correlation between FSD and economic growth. However, until fairly recently it has not been clear which direction this causality flows in. That is, as countries grow richer does this cause their financial sector to develop? Or, does the development of the financial sector itself cause countries to grow richer (i.e. economic growth to increase)?
The World Bank has undertaken extensive research into this issue since the early 1990s, largely in the form of the cross-country regressions described above. The results suggest strongly that FSD does indeed cause growth rates
to rise – though this does not mean that economic growth does not also cause financial sector development as a positive feedback.
The size of the banking sector and size and liquidity of the capital markets are both strongly correlated with growth in GDP and can be shown to ‘cause’
this growth.14 These findings support earlier work by King and Levine (1993), which looked at whether the degree of financial depth15 that existed in the 1960s could predict the level of economic growth, capital accumulation and productivity advances over the following thirty years. The authors find that this is indeed the case, and that the correlation between the level of initial financial depth and the later level of national income is highly significant.
For example, financial depth in Bolivia in 1960 was 10% of GDP, com- pared to a developing country average of 23% at this time. The study finds that if Bolivia had exhibited this average level of financial depth in 1960, the country would have grown 0.4% faster per year, resulting in a GDP 13%
higher than was actually the case in 1990.
Calderon and Liu (2003) study the period 1960–1994 for 109 countries and also find a strong causal link. However, while their study suggests the direc- tion of causality runs both ways, financial sector development has a stronger effect on growth than does growth on financial development, particularly in developing countries. Indeed, the authors find that causality in this direction accounts for 84% of the total relationship, against 57% in industrialised economies over a ten-year period.16
Although this work is very much ongoing, the evidence gathered to date strongly suggests that financial sector development has a real impact on eco- nomic growth rates, as modern growth theory predicts. Furthermore, this impact appears to be significantly greater in developing than developed economies.
What is the evidence on financial sector development on poverty reduction, though?
2.4.2 Financial sector development and poverty reduction
Both theory and empirical evidence suggest that financial sector development can reduce poverty via two separate channels:
1 indirectly, through its positive impact on economic growth rates, which in general terms causes a reduction in poverty levels, though, as we have seen, the extent of this impact may vary widely;
2 directly, through providing financial services to the poor.
On this second point, the UK’s Department for International Development (DFID) has the following to say:
By enabling the poor to draw down accumulated savings and/or borrow to invest in income-enhancing assets (including human assets, e.g.
through health and education) and start micro-enterprises, wider access to financial services generates employment, increases incomes and reduces poverty . . . By enabling the poor to save in a secure place, the provision of bank accounts (or other savings facilities) and insurance allows the poor to establish a buffer against shocks, thus reducing vul- nerability and minimising the need for other coping strategies such as asset sales that may damage long-term income prospects.17
In addition to these positive impacts that financial sector development can have on poverty levels, it is important to stress that the opposite is also true. Just as the provision of financial services to the poor can directly reduce poverty, the absence of these services is an important component in perpetuating high levels of poverty.
At the macro level there is also the issue of financial crises in developing countries. It is undoubtedly the case that any financial system – no matter how well developed and sophisticated – could fall victim to such a crisis under certain conditions. However, it is equally true that the robustness and level of development of the financial system also plays a key role in preventing (or facilitating) the onset of such crises. Furthermore, the extent of financial sector development will also strongly affect the transmission, depth and duration of any crisis should it occur.
This is an extremely important point, as demonstrated by evidence that currency and banking crises have reduced the average income of developing countries by 25% in the last quarter of the twentieth century (Eichengreen, 2004).
Finally, research by Jalilian and Kirkpatrick (2002) which attempts to quantify the impact on poverty levels of financial sector development finds that each positive unit change in financial sector development increases the growth prospects of the poor in developing countries by 0.4%.
Concluding remarks
In this chapter we have explored the relationship between economic growth, financial sector development and poverty reduction. Our end point has been the impact of financial sector development on levels of poverty, and our starting point was the relationship between economic growth and poverty.
As you progress through this book this theme will be stressed at each stage:
the ‘end’ that we are trying to achieve is poverty reduction – and ultimately elimination – in developing countries, coupled with high and rising living standards and a good quality of life.
We have seen that growth – to a greater or lesser extent, depending upon many individual features of developing countries – is generally associated with reductions in poverty levels. We have also seen that financial sector develop- ment is positively associated with higher rates of growth, and therefore ultim- ately lower levels of poverty. Furthermore, we have seen that financial sector
development can directly improve the lives of the poor by providing access to financial services. This much we know.
However, the chapter has also contained many caveats about the some- times shaky foundations of this ‘knowledge’, and stressed the importance of avoiding simplistic, one-size-fits-all policy recommendations in the developing world.
What we know provides a guide to what can be done. However, there are – and surely will be again – countries that have prospered by taking an unorthodox route, and all too many that have tried diligently to implement orthodox policies, only to see growth fail to take off as predicted.
With this in mind, and with our end goal clearly defined, we are ready to take the next step in exploring the world of development finance.
3 Financial repression, liberalisation and growth
Introduction
In the previous chapter we examined the relationship between economic growth, financial sector development and poverty reduction. We saw that the relationships between these factors are far from straightforward: in general, economic growth will reduce poverty – though the strength of this relationship will vary significantly – and, in general, financial sector development (FSD) will lead to higher rates of economic growth, though again the strength of this relationship will differ markedly from country to country.
We also explored some of the theoretical foundations of these relationships and saw that FSD can affect growth – and hence poverty reduction – via a number of direct and indirect channels. Fundamentally, however, the rationale has not changed significantly since the work of Joseph Schumpeter at the beginning of the last century: a well-developed financial sector mobilises and intermediates surplus capital (i.e. savings) and allocates this capital to its most productive uses within the economy. The result? Higher productivity and higher rates of economic growth.
For Eatwell (1997), these effects are derived from a combination of two theoretical positions. The first of these is the ‘fundamental theorem of welfare economics’, which states that competitive markets produce ‘Pareto optimal equlibria’. The second is the efficient market hypothesis that was introduced in Chapter 1, and which states that financial markets use information effi- ciently and allocate resources optimally. We also saw, however, that the efficient market hypothesis, at least in its strong form, is certainly not uni- versally accepted, either theoretically or empirically. Furthermore, while it can be demonstrated that competitive markets are Pareto optimal in a theor- etical sense, the ‘perfection’ of the competition required to produce this result is not a phenomenon that is seen in the real world.
However, if we put these caveats to one side for now – though we return to both in this chapter and later in the book – it is clearly the case that, even in theory, the financial system can only perform this role if prices accurately reflect underlying economic fundamentals, thereby allowing optimal alloca- tions to be made on the basis of accurate price signals. As we saw in Chapter
1, however, the tenets of the ‘efficient market hypothesis’ have been regularly challenged, not least by those who point to the evidence of market move- ments and their relationship – or lack of it – with underlying economic fundamentals.
Box 3.1 Irrational exuberance
The phrase ‘irrational exuberance’ first entered the lexicon in December 1996, when it was used by the then Chairman of the Federal Reserve, Alan Greenspan, at a speech to the American Enterprise Institute.
Greenspan posed the question of how was it possible to know whether asset prices had been driven up to unjustifiable levels by a period of irrational exuberance?
Theoretically of course, this should not happen. Friedman’s ‘rational arbitrageurs’ would step in to sell overvalued assets or buy undervalued assets, thus ensuring the prices remained at or near their equilibrium
‘fair value’. In practice, however, the facts suggest otherwise: as Robert Schiller has pointed out, fundamentals do not change to anything like the same extent as do the asset prices that are supposed to reflect them. Although the immediate aftermath of Greenspan’s comments saw stock markets fall sharply round the world, they were soon to recover – notwithstanding the Asian crisis of 1997/8.
Indeed, this was the time when the ‘dotcom’ bubble began to really get under way, which would appear to represent the greatest example of irrational exuberance we have seen in recent times. However, as was discussed in Chapter 1, this is certainly not a new phenomenon: from the ‘tulipmania’ of the seventeenth century (where the price of tulip bulbs rose by twenty times in a month, culminating in the sale of one bulb for nearly US$80,000 in today’s money) to the South Sea Bubble in the eighteenth century (where Isaac Newton lost £20,000 and commented that, ‘I can calculate the motions of heavenly bodies, but not the madness of people’).
The psychology of such episodes is fascinating and suggests that we are not always the rational calculating machines that orthodox eco- nomic theory would suggest. As was discussed in Chapter 1, researchers in cognitive psychology and behavioural finance have shown that, in reality, people use simple rules of thumb when making their decisions – that an upward price trend will continue, for example – rather than carefully weighing the evidence. In normal times, this approxi- mates to basing decisions on changes in underlying fundamentals, but once a ‘mania’ gets going it is both difficult to stop and, as history teaches us, very difficult indeed to avoid the temptation to get on the bandwagon.