Thursday, October 8, 2009

Weaker dollar, stronger equities (4)

This chart continues to be fascinating. It basically shows that since the beginning of last year (but not before) there has been an enduring a negative correlation between the value of the dollar and the U.S. equity market. The dollar peaked in early March '09 at almost the exact time that the equity market hit bottom; since then the dollar has dropped 15% versus other major currencies, and equities have rallied almost 60%.

I've said before that I see only one reasonable explanation for why this relationship exists, and it revolves around the dollar's role as a safe haven for the explosive increase in money demand that followed in the wake of last year's financial crisis and the massive increase in government spending that was being proposed by the Obama administration. As confidence collapsed and fears of a global financial meltdown and depression rose, the dollar and dollar-based instruments such as T-bills and T-bonds became the destination of choice for those seeking security. Dollars were sought after and figuratively stuffed under a mattress for safe-keeping. As a result, the velocity of money collapsed, and global demand fell off a cliff. This year, beginning in early March, the process began to reverse. As it became apparent that the banking system was going to avoid a meltdown, and Obama's initiatives were facing serious challenges, dollar stockpiles began returning to circulation; as they were taken out from under the mattress, dollars were exchanged for other currencies and for risky assets of all sorts. As spending gradually ramps back up, economic activity is rising and confidence is returning, and it all becomes a virtuous cycle.

So the decline in the dollar's value has been driven by a rise in confidence and a rise in the velocity of money, which in turn dramatically brightens the economy's prospects; that's why equities are rising. 

This is not to deny alternative theories. One popular theory revolves around the "carry trade." The carry trade, which involved borrowing dollars and buying real estate, commodities, and other currencies, was all the rage up until a year or so ago, when it became apparent that real estate prices were collpasing. Falling asset prices forced the unwinding of carry trades and that created a vicious cycle that drove prices to extremely low levels that threatened the solvency of the global banking system. Now, with the Fed having pumped enormous amounts of reserves into the system and promising to keep short-term borrowing costs very low for a considerable period, the carry trade is being put back on. People are once again borrowing dollars and buying risky assets and other currencies.

One deficiency of the "carry trade" theory is the absence of a role for confidence. The carry trade assumes that the big swing in asset prices was due first to cheap money and second to falling prices (i.e., falling real estate prices provoked the selling of all sorts of things as investors scrambled to meet margin calls, etc.). There's no room in this theory for the massive shock to confidence that resulted from the Obama administration's very real threat to dramatically increase tax burdens in order to finance ballooning federal spending. Another deficiency of the carry trade theory is that there is very little evidence of a major increase in borrowing. Indeed, most measures of bank lending and consumer borrowing show significant contractions in recent months, and the M2 measure of money supply hasn't grown at all in the past six months.

I think it's more sensible to think that the unwinding and rewinding of the carry trade was indeed a factor in all the gyrations we've seen in recent years, but that it is just another way of explaining why the demand for money was so volatile. It's a subset of a broader explanation of what was going on.

One important implication of my explanation for the relationship in this chart is that the rise in equities is based on a real improvement in the economic fundamentals. This rally is not a head-fake, and it's not about to fizzle unless something else goes suddenly awry. The economy really is improving, and this improvement could be enduring, at least for the next year or so. All of the market-based indicators of financial and economic health agree: e.g., swap and credit spreads are sharply lower, implied volatility is sharply lower, commodity prices are higher, and global commerce is rising.

Of course, the relationship in this chart cannot continue indefinitely. At some point the dollar will approach critically low levels that could trigger a renewed crisis of confidence if something is not done to address the weakness. A forever-falling dollar would be catastrophic not only for the U.S. economy but for the global economy as well; just about everyone, especially the Chinese, has a stake in the dollar avoiding a free-fall.

I believe that the key to breaking the relationship in this chart and putting a floor under the dollar will be an earlier-than-expected move by the Fed to raise its target interest rate. This will be such a boost to confidence that it will far outweigh any threat that higher interest rates might pose to the economy. So the Fed could tighten and the dollar could rise, and equities could continue to rise as well. I don't know how imminent this is, but I suspect it won't take many more months before the Fed realizes, as the Australian central bank recently did, that the economy is no longer in jeopardy and massively easy monetary policy is no longer called for. Whenever this does occur, however, it will probably be too late, as I think the seeds of a future rise in inflation have already been sown.


John O said...

Scott, I am sure you keep getting asked about the spike in the price of gold. So, how can you explain its rise in view of your post today regarding improving confidence and upward movement in equities? Is this just a bet on inflation? Momentum creating self-fulfilling prophecy?

Sisyphus said...


Great continuation of the series and commentary. I was wondering if you could publish or calculate R-Square or at least coefficient of correlation as part of your graphs, if anything, to see goodness of fit?



piefarmer said...

Sorry if this posts twice.
Thanks again for your thoughts.
Do you have any concern, or does it impaact your thesis, that the strong stock rally is nominal only (flat if priced in gold)? Kudlow lamented this on his show Thursday.

Scott Grannis said...

Evil: A quick look at the stats shows that over the past year the correlation of the S&P and the dollar is 0.86, and the R squared is 0.76.

Scott Grannis said...

John: I think I'll post a chart that answers your question. But basically the answer is that gold rising is a proxy for the dollar falling. So gold is just another way of looking at the dollar story.

Scott Grannis said...

REW: I think it is always good to compare how financial assets are doing relative to gold. The stock rally has been flat relative to gold in the past several months, but since the early March bottom, stocks have risen 38% relative to gold. Compared to the beginning of last year, stocks are down 43% relative to gold. It's a bleak picture on balance, but there has been definite progress since March.

I'll probably need to post a chart of this as well.