For a long time, countries were believed to benefit from having a large financial sector. A financial sector provides relatively ‘clean’ employment, and it is supposed to bring economic growth. The economic benefits are a fable, claims Dutch senior policy analyst Michiel Bijlsma.
Just before the grim month of October 2008, when a lot of banks collapsed worldwide, the relative size of the banking sector in the Netherlands reached a peak. That year, the balance sheets of the Dutch banks were six times larger than the total yearly production of the country, what is referred to as Gross Domestic Product (GDP). After some years of banks reducing lending and cleaning up balance sheets, this factor has now gone down to four, a rate similar to that in the year 2000.
From an international perspective, the Netherlands is no exception. In the United Kingdom and Ireland the size of the banking sector is still five times larger than the total annual production. The EU average is around three. It is not a coincidence that the crisis hit hardest in those countries that have the largest banking sector. In the Netherlands, the UK and Ireland, the governments had to pump billions into their national banks to save them. As a consequence, in the case of Ireland, Greece and Portugal, the governments had to request financial support from the EU.
Up until 2008, national supervisors did not have much objection against a swelling banking sector. No one seriously anticipated a systemic crises to the extent and depth as seen in 2008. The risk that national governments would need to save their banks and subsequently possibly collapse themselves was also overlooked. Finally, there was the general, simple assumption that a growing banking sector would spur economic growth.
‘For a long time, we assumed that there is a positive relationship between the size of the banking sector and economic growth’, says Michiel Bijlsma, Head of the Department of Competition and Regulation of the Netherlands Bureau for Economic Policy Analysis (Centraal Planbureau, CPB). ‘The plain assumption was that the larger the financial sector, the higher the economic growth resulting from it. This view was supported by scientific research.’
These conclusions turned out not to be true. Together with his colleagues Marielle Non and Clemens Kool, the CPB researcher published a survey into the scientific studies that supposedly provide evidence for this positive relationship. The paper was presented in the context of the inaugural lecture of Aerdt Houben following his appointment as professor at the UvA Faculty of Economics and Business. Bijlsma carried out a metastudy which examined 68 comparable studies. The average result of these studies is adjusted for phenomena such as ‘publication bias’. This bias refers to the fact that the results of research on a subject may systematically drift away from reality because less important studies or studies with divergent results are neglected or not published at all.
Bijlsma concluded that the positive relationship between the size of the banking sector in a country and the level of economic growth is less strong than was always believed. Therefore, economists based their optimistic view on an insufficient number of influential studies. This is a risk, argues Bijlsma. ‘One has to try to systematically take the whole range of literature into account. And this is what you do in a metastudy.’
Bijlsma cannot confirm that there is such a thing as an optimal size for the banking sector. ‘We did not research that aspect. The optimal level will depend on lots of factors, such as the degree of concentration of banks in a country, the solvency of banks and the degree of diversification of banks in a country. That is why I cannot say whether the financial sector is currently too large or too small.’
Nevertheless, Bijlsma is of the opinion that, by 2008, the situation in the Netherlands was no longer optimal. ‘Some studies show that the optimum level of lending is around 100% of GDP. Since the crisis occurred, the financial sector has shrunk substantially, from six to four times domestic product. I think this is a good development.’ Bijlsma says, ‘My advice to policy makers would be that we no longer have to aim for an ever-growing financial sector as has been the prevailing idea for so long.’
Regarding the risk factors inside the banking sector mentioned above, Bijlsma suspects there is still work to be done. ‘Many economists say that the Dutch financial sector is still too concentrated, and not diversified enough.’ In other words: a relatively small number of banks take a relatively large part of the total market and too many banks supply the same products and services.
Metastudies may generate results about relationships that differ from conclusions that might simply be based on the average of a series of studies. This means that the results of scientific work cannot be understood in isolation. ‘When you focus on the average outcome of a series of studies, you will be overlooking the fact that non-published research also has something to say about the relation between lending and economic growth. Metastudies are valuable to adjust for these kinds of divergences.’
Bijlsma emphasises that even before the crisis the positive relationship was less prominent than always thought. ‘If we had done these types of metastudies before 2008, we would also have found that it is not right to conclude that a continuously growing financial sector is always good for the economy. So it is not true that the adverse consequences of the last crisis are responsible for the way the relationship between the size of the financial sector and economic growth has changed since 2008.’
Therefore Bijlsma claims that it ‘would have been better if metastudies had also been conducted on this subject before 2008’. Bijlsma: ‘From this perspective, our conclusion comes too late. But hopefully our results will help prevent a future crisis.’
More information? Email: M.J.Bijlsma@cpb.nl.