When is the next recession going to start? It might seem like a particularly doom-laden question, particularly as the economic recovery after the financial crisis is still insipid in much of the developed world.

But knowing whether we are at the start, middle or end of a cycle is crucial for investors globally as equity and bond returns are closely linked to recessions and recoveries.

Listen to some politicians and central bankers and it can seem as if we are only at the beginning of this cycle with economies still recovering from the effects of the financial crisis. The dominant view among investors is probably that we are mid-cycle with equity markets stagnating as leading economic indicators start to peak and corporate earnings stop surprising positively as much as they did earlier in the cycle.

But, as is often the case, the most thought-provoking contribution comes from the bears who fret that the next recession could be just around the corner. Perhaps the most powerful contribution comes from Jim Reid of Deutsche Bank in his excellent Long Term Asset Return Study.

He analysed the average length of the 33 US expansions since 1854 and his findings provide salutary reading for anyone expecting markets to recover after their recent stumble. Of the four scenarios he studies, the most optimistic is that the next recession will start in August 2012 based on the average duration of these 33 business cycles. However, if one takes the median length instead, the start date jumps forward a year to this August.

Even scarier is if one excludes the three most recent cycles from 1982, three of the five longest expansions recorded. Reid argues, convincingly, that the debt supercycle since that date has caused expansions to be particularly extended. So taking the average length of pre-1982 cycles instead, the next recession should start in March next year. Worse, the median means a US downturn would begin next month. Bracing stuff indeed.

An obvious retort could be that there is little that is average about this current cycle with extraordinary action of central banks pumping liquidity into the system through quantitative easing. One possibility is that, through a combination of QE and QE2 in the US and an extended period of record low rates, policymakers have managed to inflate another expansion in the rolling debt supercycle. The worry then is that, given ultra-low rates, the power of authorities to have impact in the business cycle is restricted.

It also raises the pressure on the slow normalisation of policy being undertaken by western central banks. The chances of a policy error are very high. Much of the market focus has been on the finish of QE2 in the US next month but just as important will be the end of the zero interest-rate policy. Should growth weaken, some investors reckon QE3 could be a possibility.

One of these, Neil Williams, chief economist of Hermes in the UK, thinks western economies will stagnate until 2013. He worries that markets are failing to price in that stagnation properly as they have been more focused on inflation risk. But government bond markets have been sending out distress signals. Since mid-April, benchmark US, German and UK bond yields have all fallen by 13 per cent. By contrast, equity markets have barely budged with the S&P 500 down by less than 1 per cent in the same period. The signals may reflect the divide between bond market pessimists and equity optimists. But it also highlights how investors have conflicting views on the potential length of this cycle. Getting it wrong could prove costly.

The Deutsche Bank study shows how correctly timing asset allocation decisions around the start and end of business cycles could lead to strong outperformance for investors. Unsurprisingly, equities fell in 29 of the 33 recessions studied while US Treasuries enjoyed positive returns in 32 of them. Reid calculated that to get the best returns investors should switch out of equities and corporate credit 4.7 and 4.3 months respectively before the downturn and move into Treasuries 4.2 months ahead.

All that suggests that anybody thinking business cycles are likely to be shorter after the financial crisis might soon need to back away from the consensus long equities and short government bonds trades.

The effect could be even more dramatic on one of the newest asset classes: junk bonds. Junk bonds have only existed in a world of extended business cycles, coinciding with the debt supercycle of the past two decades. High-yield, because of its sensitivity to corporate default rates, is even more affected by turns in the business cycle than other assets.

None of this means the business cycle is about to turn. The determination of authorities to avoid a meltdown is still powerful. But investors need to be alert to the risk that the cycle could be shorter than many expect.