I looked for the most boring possible picture of a bank but then was seduced by this, which is actually sort of intriguing. It's like a Citibank branch lurking behind some vegetation, waiting to pounce. Photographer: Munshi Ahmed/Bloomberg.
I looked for the most boring possible picture of a bank but then was seduced by this, which is actually sort of intriguing. It's like a Citibank branch lurking behind some vegetation, waiting to pounce. Photographer: Munshi Ahmed/Bloomberg.

If you like your banks boring, well capitalized and also contributing to the real economy, you will be cheered by yesterday's quarterly review from the Bank for International Settlements, which contains a special feature on "How have banks adjusted to higher capital requirements?" That's a much-discussed question with at least two possible bad answers. The banks argue against higher capital requirements by saying it would force them to cut back on lending. I wonder whether higher (risk-weighted) capital requirements will lead banks to just monkey with the models for their risk-weighted assets.

But the BIS's answer is pretty solidly the good answer: Banks have increased capital ratios by retaining more earnings, without cutting back on lending or monkeying too much with the models:

For a sample of 82 large global banks from advanced and emerging economies, retained earnings accounted for the bulk of the increase in risk-weighted capital ratios over the period 2009–12, with reductions in risk weights playing a lesser role. On average, banks continued to expand their lending, though lending growth was slower among advanced economy banks from Europe. Lower dividend payouts and wider lending spreads contributed to banks’ ability to use retained earnings to build capital.

Go team. But the retained earning comes more from retaining -- banks paid out 27 percent of their income as dividends in 2010-2012, versus almost 40 percent in 2005-2007 -- than from earning:*

The ability of banks to increase their capital by accumulating retained earnings did not result from especially strong improvements in profitability. Net income as a share of assets fell from 0.71% in the three years before the crisis to 0.52% in the 2010–12 period across the banks in the sample (Table 1). ... The fall in the return on assets primarily reflected a decline in “other income”, which is calculated as a residual based on net income, net interest income and operating expenses.

So the news is like:

  • Banks are better capitalized.
  • They are not cutting back on lending.**
  • They are cutting back on "other income," which includes (among other things) sorts of income that lots of people find naughty, prop trading and so forth.
  • Shareholders, meanwhile, are getting less cash back than they used to.
  • They're also getting lower returns: Return on assets is down 27 percent, and leverage is also down (which is what higher capital ratios mean), so return on equity is down by over 60 percent.***

Hoo boy, that is not such a great story for banks to tell their shareholders: We're less profitable, less diversified (with more of our income coming from basic borrow-short-lend-long interest income), and return less cash to shareholders.

But it's a pretty good story for the banking regulators? Higher capital requirements have to come at someone's expense, at least in the short term. It's somewhat surprising that the someone has so far turned out to be mostly bank shareholders -- the people for whose benefit the banks are nominally run -- rather than, say, Small Business Borrowers or some other more politically attractive group. You can see why BIS would be happy to advertise that result.

* Here's a table of that:

** Though that lending is a bit more expensive. Net interest income, as a percentage of assets, is up:

One of the predictions about the impact of the transition to higher bank capital ratios -- that it would lead to wider lending spreads -- appears to be confirmed, though the widening was rather mild. Net interest income rose from 1.34% of assets to 1.62% for the full sample. This 28-basis point increase in the spread between banks’ gross interest earnings and their funding costs works out to 11 basis points per percentage point of increase in the capital ratio -- which is towards the bottom of the range of estimates for the likely increase in lending spreads produced by a number of studies before the crisis.

Incidentally, the report has a good little overview at the beginning of ways that a bank could theoretically increase capital; here's what they have to say about this approach:

The most direct way to do so would be by increasing the spread between the interest rates it charges for loans and those it pays on its funding. While competitive pressures may limit how much an individual bank can widen these spreads, lending spreads could rise across the system if all banks followed a similar strategy and alternative funding channels (such as capital markets) did not offer more attractive rates.

*** From 20.7 percent in 2005-2007 to 8.1 percent in 2010-2012, per Table 2. (For G-SIBs, global systemically important banks, ROE is down even more, from 22.8
percent to 7.0 percent.)