Economies of scale in banking? Negative

As an investor, I have made it a habit to avoid large banks in my portfolios whenever I can. With this simple trick, I have done very well, especially over the last decade. The problem with large banks is simply that they have in my view all become too large to manage and too large to be profitable. Because established high street banks have become so lethargic and complex, small challenger banks could gain market share (and might actually be worth an investment in some cases). 

With the rise of challenger banks, established banks tried to copy the tools and methods of these challenger banks in the hope that their economies of scale would allow them to keep market share. The problem just is, there are probably no economies of scale in traditional banking.

It is an open question if bigger banks are able to gain bigger margins than smaller banks and it is empirically hard to test if this is so. The problem in practice is that large banks don’t get large by accident. Instead, large banks are the result of a series of mergers and sheer luck and not of organic growth that outpaced competitors. That big banks did not get big through organic growth is clear whenever you look at the history of any major bank in the world. That luck plays a crucial role becomes clear when you look at who survived the financial crisis and who didn’t As Michael Lewis describes in his great book The Big Short, Deutsche Bank was lucky because it had a few traders who were willing to bet against their own fixed income structured products and thus provided an accidental internal hedge against the crisis.

But in some instances, it is possible to check if larger banks really have economies of scale. Kilian Huber recently investigated one such case in Germany. After the Second World War, the allied forces broke up the banks and only allowed smaller regional banks to operate. This was done because large banks like Deutsche Bank or Commerzbank were major financiers for the Nazi regime. However, between 1952 and 1957, the allied forces relaxed their restrictions and gradually allowed smaller regional banks to combine back to larger economic entities. These weren’t mergers to grow the business but simple consolidation measures to clean up a fragmented banking system.

This natural experiment allows Kilian to test if the merged, larger German banks did indeed become more efficient than their untreated smaller peers that didn’t get merged into larger entities by the allies.

His answers are pretty telling. First, treated banks (i.e. banks that got merged into larger entities) did not accelerate lending. If anything, the larger banks had slower lending growth to non-banks than the smaller peers – though the difference was not statistically significant. Also, the growth of the companies that used bank loans from larger banks was no different from companies that had loans from smaller banks. So larger banks weren’t able to provide more growth capital or select better lenders than smaller banks.

Lending growth of merged larger banks (treated), vs. smaller banks (untreated).

Source: Huber (2021).

But there were a few differences between the merged larger banks and the smaller ones. Larger banks tended to be more willing to lend to riskier, more volatile borrowers and were more willing to allow a borrower to lever up. In other words, larger banks were chasing profits with riskier lenders and were willing to allow these borrowers to increase leverage. This probably works well in boom times, but as we all know, these overleveraged borrowers often fail when the next recession hits. And this is probably why in the long run, the larger banks weren’t more profitable than the smaller banks.

But there was one other factor that materially differed between larger and smaller banks. Executives at larger banks saw their salaries rise 251% between 1952 and 1960 while executives at smaller banks saw their salaries rise by only 102%. And the media attention also grew exponentially. Looking at media reports on German banks in the German news magazine Der Spiegel and the FT, Huber found that “one bank of size ten receives more media mentions than ten banks of size one combined”. In other words, the executives got fame and fortune while the rest of the economy got nothing.