On Buiter and the Crisis

In August 2008, a month before Lehman collapsed, Willem Buiter presented what he light-heartedly called the "longest paper ever" at the Jackson Hole Symposium. The paper made waves around the world for primarily two things. The first was its length, a staggering 141 pages. The second, and more substantive was its criticism of the handling of the crisis (until then) by the three main central banks in developed nations (the Fed, ECB and BoE).

I finished reading the paper two days back. For me, the best and the most striking parts of the paper were not those that dealt with a judgement on how the central banks had been handling the crisis. They were the connections that he makes between macroeconomic (output, price, inflation) stability, financial stability and the central bank's role in binding the two.

Buiter begins with the premise that in any financial crisis, the job is two-fold. First, the immediate damage to the economy at large has to be minimized. Next, the probability of occurrence of future such crises has to be minimized and better tools to deal with them have to be developed. He then offers (or borrows form others and integrates) several great insights on the linkages between macroeconomic and financial stability.

1) The short term interest rate is a blunt and indirect measure suitable for macroeconomic stability but not for financial stability.

2) Reserve requirements (or Cash Reserve Ratios) are a quasi-fiscal tax on banks when they offer no interest. When they offer interest, they can be used as tools of financial stability.

3) Asset price booms and busts are always asymmetric. So is the leveraging and de-leveraging of the financial system. Booms and leverage ratios always build gradually, though they can reach monstrous proportions. Busts and de-leveraging are much more rapid.

4) Asset price bubbles are driven, by definition, by non-fundamental factors. The short term interest rate is a fundamental determinant of asset prices and is thus too insensitive to be used to prick such bubbles.

5) It is not just huge commercial banks that are 'too big to fail' . Any reasonably large or inter-connected institution that has considerable leverage and the majority of its assets as financial assets are 'too systemic to fail'.

6) Two different but inter-connected kinds of liquidity crises can lead to and feed from a system-wide solvency crisis. One is funding liquidity, which occurs on the liability side. Here, a financial institution is not able to borrow overnight or short term to meet certain regulatory or business requirements. The other is market liquidity, which occurs on the asset side. Here, a financial institution is not able to sell an asset on its books in the market to get cash as the market for that asset has frozen. Clearly, the two kinds of liquidity are inter-connected.

7) The central bank has two functions in such a situation: as a lender of last resort (LLR) to solve a funding liquidity crisis, and as a market-maker of last resort (MMLR) to solve a market liquidity crisis.

8) A special resolution regime with prompt corrective action (PCA) measures is required to ensure that fundamentally insolvent financial institutions are allowed to fail (as solvency is a private good) without damaging the rest of the financial system and the economy.

9) In the LLR function and MMLR functions, central banks should be willing to accept as collateral and/or purchase and hold on their books large varieties of assets that may not be good today but will be good of held to maturity. The discount/lending rates on such lending and purchases should be punitively high.

10) Such lending and purchases have to be buck-stopped by the treasury, ideally by an immediate exchange of risky assets with sovereign debt between the central bank and the treasury. This is to ensure that the central-bank is not abused as a quasi-fiscal institution and the risk is undertaken on the books of an institution answerable to the democratically elected legislature.

11) The secured overnight inter-bank lending market (call money market in India or the Federal funds market in the US) exists mainly due to bizarre central bank procedures and is rather redundant. To set a truly effective official policy rate, central banks must be willing to borrow and lend any amount at that rate. Because central banks have punitive rates or quantity caps on the overnight lending and borrowing that they do from banks, effective risk free rates often diverge from policy target rates. (A good example of this is the call rate in India, which has been hovering between 2% to 3% for the last few months even though the reverse repo rate is 3.25% and repo rate is 4.75%.)

12) Any financial system has inside assets (where the asset and the liability are both financial in nature) and outside assets (where the asset is in the real economy and the liability is financial). Home equity and stocks are outside assets while home mortgages and debt instruments are inside assets.

13) The modern financial system has many layers and a majority of its assets are inside assets. Even a rapid de-leveraging of such a system can be sustained if it can be ensured that the effects don't spill over into outside assets. More importantly, the counter-cyclical policy measures to flood the markets with liquidity in a crisis need to be tempered by the fact that a lot of de-leveraging may simply involve inside assets.

14) Monetary and credit aggregates are important tracking tools and obsession with the short term interest rate is counter-productive.

Apart from these, he makes some other key points as well.

1) Lack of transparency in the pricing of illiquid collateral creates moral hazard problems. When coupled with the in-built adverse selection in the way the Fed prices these collaterals - it accepts the pricing of a clearing bank that is typically that is typically a business associate of the broker-dealer in question - the US had created the mother of all moral hazard problems even before Lehman went under.

2) Due to its failure to differentiate between inside assets and outside assets, the Fed reacts far too strongly to weak asset prices. Or, it has been 'cognitively captured' by Wall Street.

3) There is a structural break in the relationship between core inflation (doesn't include energy and food prices) to predict future headline inflation (includes everything). This is due to increased consumption by India and China. As a result, Fed's monetary policy has been too loose and as the financier of the world and the dollar-country, the US has been exporting this inflation elsewhere.

4) The US and UK have current account deficits as well as low-interest bearing assets. What this means is that the US and the UK are living on borrowed consumption being financed at effectively negative nominal and real interest rates! This won't continue for a long time. A massive outbreak of inflation will follow in the middle run (2-3 years) unless there is a structural change in the ratio of consumption to savings.

5) With foreign assets and liabilities at 500% of GDP, the UK is almost like a giant hedge fund.

Buiter also shows remarkable foresight when he contends that the fed funds rate (which was 2% then) might hit the zero lower bound soon enough, and that the worst is not over. He also makes the interesting proposal of de-linking reserves from currency, arguing that bank reserves kept with the central bank can pay a positive as well as negative interest (storage costs and security costs of cash will ensure that banks will not mind paying a small interest to the central bank to keep reserves). This could then free up the policy option to actually have a negative nominal interest rate to counter deflationary pressures.

Where I fail to understand Buiter is if he supports or opposes counter-cyclical capital requirements for financial institutions (which go up in times of plenty and down in times of crisis). Early on in the paper, he recognizes leverage as the key villain and endorses the Goodhart-Persaud proposal of counter-cyclical capital and liquidity requirements and extends it to all large leveraged financial institutions from just commercial banks. He also argues that such requirements should be based on rapid leveraged balance sheet growth of the institution in question and that national regulators can and should go beyond Basel II in ensuring this. Not much later, however, he takes his oft-repeated stance of 'liquidity is a public good, solvency is a private good' and argues that any extra liquidity requirements during good times to provision for the bad times is a privately and socially inefficient waste of liquidity. He also makes the same case in this blog-post, where he trashes precisely the kind of move by the FSA that he seemed to be supporting early on in the Jackson Hole paper. If his point is that liquidity provisioning in bad times cannot solely be a private endeavour then I agree, but he seems to be making the case that liquidity provisioning in bad times should be completely a public endeavour undertaken by the central bank which can create new liquidity almost costlessly. It is hard to understand how such a view reconciles with a support for counter-cyclical capital and liquidity requirements.

Nevertheless, this is probably the only paper in which one economist manages to form a coherent whole of a wide variety of divergent thoughts on the crisis. Read it at your own leisure. It will continue to be relevant long after this crisis is gone.