Saturday, October 23, 2010

What I am reading: Hicks Classic Paper

Mr. Keynes and the "Classics"; A Suggested Interpretation - J. R. Hicks (1937)

This is the paper that introduced the IS-LM apparatus as Hicks interpretation of Keynes General Theory.  Hicks is frequently criticized for abandoning Keynes focus on the financial market speculation as a driving force behind booms and busts but the revolutionary element of this paper is pretty apparent when you compare it to what else was going on at the time.  In this paper Hicks created the framework that macroeconomics would build around for the next 40 years.  He does this by constructing simple equilibrium models representing what the "Classicals" (read Pareto, Marshall, Pigou) had argued and what he interpreted Keynes to be saying (at least in part) in the General Theory.

Hicks assumes in both models an IS equation representing equilibrium in the market for loans i(r)=I = S(Y,r).  He then argues that the major difference between the Classicals and Keynes is in the specification of money demand.  The Classicals attribute money demand to a transactions motive as represented by the Cambridge equation M=k*Y while Keynes argues that money is held as an asset - in simplest form M=m(r).  Hicks extends this to the "General" theory M=m(r,Y) which is our modern LM equation.  Note that in the Classical model M -> Y without any involvement of r.  r and I are then determined by Y and the IS equilibrium.  In Keynes shocks to M are passed through to Y at least in part via M's impact on r.  Keynes simple case says M->r->Y and Keynes general case says M->(r,Y) two equations in two unknowns which must be solved simultaneously.

Finally Hicks asks how we are to reconcile this model with Keynes view that monetary policy was ineffective in escaping major downturns.  Hicks posits the liquidity trap - an L shaped LM curve with a horizontal portion on the left and a vertical portion on the right.  If the IS-LM equilibrium (r,Y) currently lies on the horizontal portion of the LM curve then manipulating the money supply shifts the LM right / left but leaves the (r,Y) equilibrium unchanged.  Fiscal policy however will shift the IS along the horizontal LM leading to a change in Y while leaving r unchanged.   If the current IS-LM equilibrium (r,Y) currently lies on the vertical portion of the LM curve then monetary policy will again be effective.

Next up Modigliani...

Sunday, October 10, 2010

I am reading: All About High-Frequency Trading

All About High-Frequency Trading - Michael  Durbin

If you want to know what high frequency trading is all about and what some of the issues are surrounding high frequency traders then this is a nice book.  If you want to set up your own high frequency trading firm this book is not going to get you too far - although perhaps that is too much to ask from any book.

One of the questions touched on this book is - are high frequency traders good or bad for the industry?  That is not a well framed question.  High frequency trader refers to a large class of  traders running many different strategies, and their good or badness is dependent on the particular strategy employed.   The book describes some of the strategies that high frequency traders may use.  Some of these strategies are versions of market making (clearly a good) while other strategies profit at the expense of customer orders (arguably a bad).  Some strategies are stabilizing while others can be destabilizing (read trend followers).

I work with a number of high frequency traders and in general I see them as a positive force for the industry as they provide liquidity into the markets and keep price relationships in line with fundamentals.  Large orders can then come into the market without causing significant price disturbance.  However there is always the fear of the single runaway black box or of black boxes getting stuck in a self reinforcing negative loop.  And the speed at which these negative effects could potentially come into play is concerning.  This is what the industry is still trying to get their arms around.

Saturday, September 18, 2010

I am reading: Would a stock market drop affect consumption?

Wealth and Consumption: Would a Stock Market Drop Really Cause a Recession - Stephen Cecchetti (2000)

A few facts 1990-2000
  • US household wealth doubled to $44 Trillion
  • the value of the stock market increased 300%
  • household consumption increased 60% to $6.2 Trillion
  • the ratio of savings to disposable income fell from 7.8% to 2.4%
Cecchetti notes that post 1995 the ratio of consumption to wealth dipped below the normal range of 17-21%. By 2000 from the perspective of a consumption to wealth ratio people were consuming as though the stock market were 25% lower that it actually was. This may have been due to a time lag to adjust consumption or uncertainty about the state of the stock market. He hypothesizes that a 25% drop in the stock market should have little impact on consumption - unless it generated a panic.

I am reading: Aggregate wealth and aggregate consumption

The Wealth Effect in Empirical Life-Cycle Aggregate Consumption Equations - Mehra (2001)

main results (approximately):
  • cointegration / error correction model.
  • in long run a $1 increase in annual disposable income -> $0.55 increase in annual consumption.
  • in long run a $1 increase in wealth -> $0.04 increase in annual consumption
These are not stunning revelations as they confirm previous results. However it is something to think about given the size of the swings in the stock and housing markets that we have experienced over the last decade.