The consensus view among investors from specialist commodity fund and hedge fund firms is that commodity prices are likely to rise in the coming decade. As stock pickers, they believe valuations in many commodity-related companies may therefore still be cheap.

This may be a surprising conclusion, given the sharp rise in most commodity prices since 2001, which have this year broken through highs not seen since the early 1980s. But investors such as Darko Kuzmanovic, long/short equities portfolio manager at the Prudent Natural Resources Fund in Canada, says the asset class is at the early stage of a long cycle of capital formation.

The three decades from 1970 to 2000 saw demand for commodities at "GDP-minus", ie prices rose less than GDP growth rates. But since 2000, global demand has risen at "GDP-plus", around 4-6% CAGR. Therefore in a short space of time, prices in commodities such as copper have risen from 65 cents an ounce to $2.20, while Brent crude prices have gone from $30 to $70 a barrel.

Kuzmanovic says some markets such as copper are now stretched and due for a correction, but ultimately demand will continue to increase.

And this demand is not just the China story, says Klaus Rehaag, executive vice president at Gardner Finance, a Swiss-based fund of energy and commodity funds, and a private-equity and structured finance specialist. "The United States is the other big source of demand growth. The trillions of expenditure needed in energy is to replace old, rusting assets in the West. Electricity grids in the US and Europe are 30 years old."

Which leads Rehaag to the point that supply problems will continue to contribute to rising commodity prices. Future resources come from further afield, in smaller amounts, in more difficult conditions for extraction. And whereas for the past three decades, most of the world's mining supply came from the United States, Canada, Australia and South Africa, today most energy and metals resources come from places such as former Soviet Union states, Latin America and the Middle East. "Resources in Russia and Saudi Arabia are difficult to extract even for companies with plenty of cash," he notes.

Commodities experts favour the argument that a full cycle lasts around 30 years, the time it takes to not only explore for new resources and build a refinery but also to obtain the government licenses and permits required. Governments such as Venezuela's are too unpredictable for most global companies, while others such as Mongolia's can haggle for years over new finds. And the virtual end of exploration in the 1990s has left most areas of mining and energy without an existing pipeline of large-scale projects.

Kuzmanovic adds that on top of supply and demand, costs in many industries such as mining are also rising. Again, a lack of investment in the 1990s means suppliers of specialised equipment such as trucks and tyres can't expand capacity quickly enough to keep up with demand.

Skilled labour is another problem. The malaise of mining and energy industries in the 1990s saw most engineers and other professionals retire. The average age of these people is in the mid-50s. These people are not keen, therefore, to head back into the field in Nigeria or Mongolia. That's a job for younger people, but there are fewer younger people with the requisite skills, even for positions that seem straightfoward, like truck drivers. So there are many structural factors behind rising costs of extraction and, therefore, of commodity prices.

Michael Coleman, Singapore-based managing director at Aisling Analytics, which runs a commodities fund that trades via derivatives exposures, says that current market valuations for many commodity stocks don't reflect these realities. The market tends to price oil stocks, for example, at $40/barrel, not $60. With oil prices above $60/barrel, therefore, these stocks appear pricey. But if you believe in the long-term cycle, these stock prices should in fact rise. Eventually the underlying commodity price assumptions will also have to rise, to keep up with reality.