Monday, March 31, 2008

Banking and currency crisis – there is a link


A well-established relationship is the link between banking and currency crisis. While we have been through a period of calm in developed countries over the last ten years, an application of the twin crisis concept suggests that these negative events are linked. Glick and Hutchison, two researchers from the San Francisco Fed, have reviewed the link between banking and currency crises in 1999 and found that most banking crises precede currency crises, but the opposite is not true. Currency crises do not always lead banking crises. For the period sampled (1975-1997), the researchers found that of the 90 countries looked at 72 had banking issues and 79 had at least one currency crisis. The period saw 90 banking and 202 currency crises with the number increasing in the 90’s. Twin crises occur more often than many may anticipate. My reading of the data suggests that the prevalence of crises has fallen over the last 10 years but has not been eliminated.

While most of the crises have focused on the emerging markets, the same logic could apply to larger developed countries like the US. Because banks are not as important to the overall capital markets in the US as in developing countries, we would have to broaden our definition of banks to include other financial intermediaries or the shadow banking system. Many banking crises are caused by bubbles like what has occurred in the US mortgage market. Clearly, the values for homes have increased at an alarming rate relative to the historical and some valuation techniques. Banking crises are also tripped by non-fundamentals like a bank run which occurred with Bear Stearns. Finally, banking crises can be caused by moral hazard problems associated with financial liberalization. The whole development of new mortgage products in an unregulated manner may proxy for liberalization risk. http://www.frbsf.org/econrsrch/workingp/pbc/1999/wppb99-07.pdf

The issue is whether we can classify the current credit problem in the US as a banking crisis and whether the sell-off in the dollar is a currency crisis. I would argue that we are in the midst of a twin crisis, even if the criteria used to measure these twin crises in emerging markets are not met for the dollar.

Banking distress, like we are currently facing, will have an impact on capital flows. The dollar is fundamentally different as a reserve currency hence the change in the exchange rate or the measure of a systematic banking problem may be muted. The channel of transmission for the currency crisis in this case is not a failure of banks which creates risk aversion with investors but a monetary policy stance by the Fed which is out of synch with the rest of the world. The Fed is easing fast in a inflationary environment when other central banks have been tightening, This difference in monetary policies is consistent with “first generation” model of currency crises based on fundamentals; however, the change in risk aversion of many global investors have portfolio rebalancing effects on the dollar which is consistent with “second generation” models of currency crises.

Most central banks have started to tighten monetary policy except for those countries which have been affected most by the credit crisis. The three central banks which have eased are the Fed, Bank of Canada and the Bank of England. All are facing different degrees of credit/financial crises. For Canada, the link with the US has forced easing as well as the crisis associated with commercial paper funding. For Great Britain the problem has been with mortgage lenders which have lead to the take-over of Northern Rock.

The twin crisis framework may provide a good way of looking at how the dollar may trade for the rest of the year. Any improvement in the bank crisis will carry over for dollar strengthening.

No comments: