US stocks have displayed a positive correlation with bank reserves (i.e. banks’ account balances at the Fed) since QE1 was launched in late 2008. A post on 18 September noted that the Dow Industrials index, then at 13,553, was 1,300 points above the level implied by the current level of reserves, based on the historical relationship. This suggested that QE3 was priced in.

Subsequent market weakness resulted in this gap closing to 970 points by the time of an update post on 30 October. The Dow, moreover, was then slightly below the level implied for end-March 2013 assuming continued unsterilised Fed bond purchases of $40 billion per month and no offsetting changes to its balance sheet. A buying opportunity, in other words, seemed to be developing, especially if the Fed was thought likely to boost its bond buying when “operation twist” expired at year-end.

With the further fall this month, the Dow closed last week 450 points above the level implied by current reserves but 570 points below an end-March “forecast” assuming QE3 continues at its recent pace and 1,160 points lower than a prediction based on bond purchases stepping up to $85 billion from year-end – see first chart. The relationship, that is, suggests a rise in the Dow of between 5% and 9% by next spring. Last week’s dovish October minutes have shortened the odds that the Fed will announce an expansion of bond buying at its 11-12 December meeting.

A market rally between late 2012 and spring 2013 would maintain the similarity with movements 25 years ago, albeit that recent fluctuations have been smaller in magnitude* – second chart. The Dow rallied 19% from a low on 4 December 1987 to a short-term top on 12 April 1988, working its way gradually higher over the remainder of that year.

*A post in May noted this resemblance and suggested that it would continue.