04 September 2013

Banks: Collecting the Rent

Since the start of the crisis I have been trying to understand how it is that banks (and the finance sector more generally) extract rent from the economy. I once quoted Simon Wren-Lewis asking the right question:
Did innovation and deregulation in that sector add to social welfare, or make it easier for that sector to extract surplus from the rest of the economy?
To explain how far I've got in my thinking, I will need to make some gross simplifications about how banks work. Bear with me, it help to gain an insight into the subject.

The first step is to see that banks profit margin comes from the difference between the interest they pay (to depositors or bond holders etc) and the interest the charge to borrowers. Out of this margin they pay their running costs and the rest is net profit.

The next step is that banks make money from all the loans they have issued, which are recorded as assets on the banks balance sheet. So as a first approximation its rate of profit can be seen as its total net profit divided by its total assets:
profit/assets 
 
Banker however do not pay much attention to this ratio. They are more interested in maximising the return on equity, that is the highest profit for the shareholders' capital.
profit/equity
 
An interesting analysis piece in the FT explained why bankers have an incentive to watch ROE:
...return on equity... is a benchmark that investors use to compare stocks and it is also used in the calculation of bankers’ bonuses.
Now comes the clever bit. We can analyse ROE with this equation:

 profit/equity = (profit/assets) x (assets/equity)
 
The first term in the equation is return on equity (ROE). The second term can be considered to be the rate of profit on its loans or "profit margin". The final term is a measure of how much the bank funds itself through debt rather than equity. So the equation says:

ROE = profit rate x leverage
 

Here is an example. Before the crisis, a high street bank might have had profits of 4 billion on assets of 800 billion and equity of 20 billion. So its ROE was 20%, its profit margin was 0.5% and its level of equity at 2.5% of assets gives a leverage of 40:

20% = 0.5% x 40
 
Notice first how tiny the rate of profit is. For a non-financial company its total profit is the rate of profit times the amount of sales. So to make a decent level of profit banks need to sell lots of loans. That is what happened. Banks turned from being prudent institutions safeguarding our money and making loans to good customers to become selling machines pumping out loans to the next mark customer.

Next look at the level by which ROE is boosted by leverage. A small profit is boosted to a large profit (40 times larger) by using debt. If the bank increased its equity from 2.5% of assets to 4% of assets, it leverage would be 25. In the example above its ROE would fall to  about 12.5%. (Actually not quite as it would need to pay less in interest on the equivalent of 1.5% of assets so its rate of profit would rise slightly.)

The problem is that this level of leverage is dangerous. If it makes loses in one year equal to 2.5% of assets then the equity is wiped out and the company becomes insolvent. For any other type of company it would drive up its cost of debt. Banks by contrast have had an implicit guarantee that the state would come to the rescue.

This then is how banks extract rent, by expanding the amount of loans they make beyond what is prudent, and using debt to magnify the return on equity. In the process they create a highly unstable economy where the public ends up paying the bill.

Two final points. The equation profit/equity = (profit/assets) x (assets/equity) is one of the three equations to explain the crisis that I promised some time ago. The others will follow later. Secondly, I've avoided discussing the Miller-Mogdliani thoerem, partly to keep it simple and partly because I want to post a review of Admati and Helliwig's The Bankers New Clothes before getting into that one.