A popular example of the complexity of "chaos" theory involves a butterfly flapping its wings somewhere in Latin America and through an intricate chain of linkages setting off a powerful tornado thousands of miles across the globe. The current financial crisis is perhaps an equally poignant illustration. What started as a sell-off in the market for American mortgage-backed bonds sometime in July 2007 has now manifested in a sharp rise in inflation rates in Asian economies.

One of the errors in judgement, in my opinion, that economists made at the early stages of the crisis was in making the following assumption — that the transmission of the US crisis would work essentially through macro-economic linkages. The key elements of this hypothesis were the following. Falling house prices would erode household wealth and destroy consumer demand. Hurt by a rising share bad loans on their books, banks would cut back on credit. This would choke both consumer and investment demand further. As the US slumped, the rest of the world would also slow down since trade and investment flows with the US are really what kept the rest of the world ticking. This seemed like a fairly neat hypothesis except for one problem. The corollary of this sequence of events would be a sharp decline in all commodity prices as global demand contracted. Thus the fact that oil sells for close to $150 a barrel today and steel prices are soaring suggests a fundamental flaw.

Where specifically did this hypothesis go wrong? Has the US showed more resilience to the problems in its mortgage market than expected? Not quite. There is enough evidence that the US has slowed down substantially over the past six months although it is technically perhaps still not in recession. Consumer spending has dissipated and corporate sales are sluggish. The housing market is in free fall. The US may be in deep trouble but has its problems impinged on the rest of the world? This is where the macro-transmission hypothesis begins to falter. The rest of the world does not seem to have been affected as much or as quickly as many had anticipated. Europe, for one, is still holding up though there are growing signs of moderation. Asia certainly has not collapsed.

Does this mean that other economies have entirely decoupled from the US growth engine? The jury is still out on this. It is possible that it is just a question of leads and lags. Other economies will respond to the US with a longer lag than anticipated by the end of this year, the global economic system would have slowed considerably. However, at least for me, this is a somewhat peripheral issue or question.

What seems more important is the fact that the initial shock (that emerged from the US mortgage markets) has transmitted to the rest of the world not through visible macroeconomic linkages but rather through dramatic shifts in risk appetite and asset preferences. Thus, the first signs of an end to the crisis have to be found in asset market movements, rather than in hard macroeconomic data.

Let me explain. By the beginning of the year (that is, roughly six months into the US mortgage crisis), most analysts had recognised that risk aversion had risen sharply and the search for a safe haven would determine the way investors would allocate their assets in 2008. Emerging markets, touted as a hedge against the US for a while last year, soon fell out of favour. Analysts predicted that global investments would find their way into the time-tested "safest of the safe" havens, US treasury bills and bonds. Thus, despite the underlying problems in the US, the flight to safety would support the dollar.

This, however, was not to be. Two alternative assets classes emerged as a viable hedge against the dollar. Investors seeking insulation from the heat of the US economic meltdown started to hold commodities (particularly oil) and short-term European government debt instruments. As the flow of bad news from the US accelerated, investors dumped dollar assets and bought heavily into these asset classes. Both the euro and commodity prices gained as a result. There were other quick intermediate effects — rising oil prices fed the expectations of an accelerated shift of cultivable land away from food into bio-fuels. Food prices rose in response.

The net result was the emergence of a global macroeconomic crisis that involves a peculiar conjunction of two trends that currently run parallel but are usually in conflict with each other. The prospect of a global slowdown that going by conventional wisdom should breed deflation goes hand in hand with rising inflation. This makes it so difficult to handle. Central bankers find it difficult to figure out which of the two evils is lesser. Do they pare interest rates to thwart a slowdown or raise them to fight inflation?

My sense is that the beginning of the end of the crisis will be seen when this dilemma is resolved. Just as the transmission of the crisis worked through asset markets, the resolution will also lie in the asset markets. Renewed preference for dollar assets will lead to moderation in commodity prices and will go hand in hand with euro depreciation. As this happens, inflationary pressures will wane. Central banks will then have a more defined remit — to get growth back on track. How soon can this happen? As soon as investors get the first whiff that the US business cycle is close to a bottom, risk preference could change dramatically. That is certainly possible in the next few months.

The author is chief economist, HDFC Bank. The views here are personal