Godzilla and the Aspen tree
06 November 2012
By Nick Forrest
As part of the PwC-sponsored Beesley lecture series, Andrew Haldane, Executive Director of Financial Stability at the Bank of England gave a fascinating lecture titled “On being the right size”. He started with an odd one out contest. Lining up were Godzilla, King Kong and Santa Claus. Andrew revealed Santa was the only character who is physiologically possible – Godzilla and King Kong are so large, that, if they did exist, their skeletons would be unable to support their weight. They are physiologically impossible. This was some introduction to a talk about the size of banks.
Andrew’s hypothesis is that banks have been supported by implicit (and sometimes explicit) guarantees, which have given them incentives to grow. The larger they grow, the more they benefit from government support and the more profitable they become. Andrew’s empirical analysis supports this suggestion that bigger banks are more efficient. However, when he adjusts funding costs by stripping out the benefit of this government support he finds that banks become less efficient above a size of $100bn in assets. By way of comparison our biggest banks have over $2tr in assets, so this would imply a radical restructuring in the shape of the banking industry. This finding, he suggests, helps to explain why banks have a commercial incentive to expand, but beyond the point which is in the public’s best interests.
So is the benefit of implicit government support so great? Andrew suggests it may be worth as much as $700bn per annum across the 29 largest global banks. What is clear is that banks benefit from implicit taxpayer support as measured by the difference between their stand-alone and supported ratings, which usually translates into lower debt funding costs. This gap may have narrowed in the UK, but it is still present in current bank credit ratings, as can be seen below.
[click image to enlarge]
What is hard to assess is who benefits from this implicit government support. Current returns across the banking industry are sub-10%, with a cost of equity of around 11% (see risk and return in the banking sector), so bank shareholders are not currently earning excessive economic returns. If we go back to 2007, the last year before the crisis, and take Barclays as an example, it earned a ROE of 16% on an equity base of £32bn. If we assume that its cost of capital was around 11%, then this 5% difference (worth about £1.5bn) could be attributed to a shareholder benefit derived from implicit government support. It is hard to see how we’ll get to $700bn on this basis.
If bank shareholders have not benefited, then who has? Bank employees may have earned more than they would otherwise and there is a good case that financial markets have attracted talent from the rest of the workforce, through high salaries. The total staff costs of 24 of the 28 largest global banks is around $300bn, but many banking sector jobs have comparable jobs in other sectors, where pay levels should equalise. It will be hard to tell how much of this is inflated, but it would still be a long way short of $700bn.
So the final beneficiaries from implicit government support could be customers. In a competitive market, any cost advantages should be passed onto customers through lower pricing (or an increased availability of credit to more risky borrowers). This is the more plausible outcome, but does lead to difficult policy questions. How do we achieve a lower cost and wider availability of credit to provide essential finance through the economy (the positive externality) and at the same time prevent the bubbles which can lead to crises (the negative externality)? A complex mix of macro and micro prudential regulation is required.
But even if bank shareholders have not benefited greatly from implicit government guarantees, banks may still be too big to manage, regulate and value. What does the market think? The analysis below sets out the market price to book equity value for banks in different size buckets. It would appear that larger banks are associated with lower relative market values, but there will be other factors to consider here (such as the larger banks being exposed to LIBOR-related fines and litigation). The market perspective generally supports Andrew’s position (albeit at a higher size threshold).
So where next for the big banks with both regulators and the market challenging their size? One response is to shrink and many banks are disposing of non-core assets, but the other response is to simplify. This is because there is nothing wrong with big companies per se. Apple is a successful, large company based around a small number of products. To return to analogies from the natural world, the largest organism in the world is arguably the interconnected root system of the Aspen tree. The largest known aspen grove is located in Utah and covers 106 acres and is estimated to weigh 6000 tonnes. So you can be big as long as you are not too complex.
Nick Forrest | Telephone: +44 (0)207 804 5695