The auto industry has been one of the hardest hit by the collapse of credit markets and consumer confidence since last Summer, and given its huge weight in many economies such as Germany and Japan, and key role in determining many commodity prices from platinum to copper, the scale of any recovery is crucial for investors. Complicating any analysis are the numerous national incentive schemes designed to artificially boost short-term demand (by about 250k units in the US alone this year) and the growing role of China in the global industry. Over the past 5 years, Chinese car production has doubled (almost entirely for the domestic market) and will become the world's largest producer in 2009, while US production has halved. Indeed, the collapse of the US car market from a 14-17m unit average since the early 1990's to recent annualized rates of 7-8m has been the most spectacular worldwide. Currently, Chinese car sales are booming (up over 40% yoy), notably of the smaller, most fuel efficient models (that benefit from government tax incentives and 'no collateral' loans) and in which Beijing is attempting to build critical mass for an export push following Japan's development model. As with all temporary incentive programs, there is a risk this is merely distorting underlying demand by front loading purchases, and is unsustainable.

The global industry went into this crisis with huge structural overcapacity, in the order of 25-30m units a year. However, in most countries inventories have now been slashed to levels that suggest a sharp rebound in production is likely through early 2010, and US production should rebound to an annual rate of more like 10m by end year as pent-up demand from the natural replacement cycle and corporate/rental fleet sales kick in. Steadily rising prices for used vehicles are one positive lead indicator for this scenario, and it would have bullish implications across many sectors from rail shippers to sheet steel producers, and indeed for US GDP growth to turn positive.  The chart below looks at projections for production in the key regions over the next few years.

These figures, from industry consultants iSupply, probably underestimate the 'dead cat' move likely in the next 12 months but overestimate the medium term recovery in my view. Make no mistake, the durability of a recovery should be seen in the light of the 'overconsumption' of autos as with housing and other consumer durables in the boom years. We won't see anything like 14m autos sold in the US again for several years in a slow growth, high unemployment and tight credit environment. Nonetheless, having undershot on this as so many other economic indicators, we will level out in the next 6-12 months and industrial stocks will benefit. There are a variety of ways of playing this rebound from auto shippers like CSX to Ford which has navigated this crisis admirably well (post a likely fund raising) and platinum (either via the metal ETF/ETN or pure play miners like Lonmin), which with its key role in auto catalysts, is a highly geared play on production volumes.