Yesterday's big stock market falls pose the question: do we really have anything to worry about?

It's tempting to answer: no. The claim that stock markets have predicted nine of the last five recessions is a cliche because it is true - although why (pdf) this is is a matter of doubt. And we know from the very mild repercussions of the tech crash of the early 00s that exogenous falls in share prices don't have catastrophically negative wealth effects.

Nevertheless, I suspect there are four concerns here.

First, the fact that the Chinese economy is slowing down (pdf) is a symptom that world trade generally is stagnating; although today's figures from the CPB showed a good increase in this in June, the volume of world imports is still lower than it was last autumn. This means that a major force for the world's enrichment - an increased global division of labour - is now absent. Unsurprisingly, there's a close correlation between annual growth in world trade and annual returns on UK shares.

Secondly, we don't know how much further the world economy might weaken. Granted, the stock market is a lousy predictor of recessions. But so too is pretty much everything else (except perhaps the yield curve). Back in 2000, the IMF's Prakash Loungani wrote (pdf) that mainstream forecasters' "record of failure to predict recessions is virtually unblemished": the recession of 2008-09 merely corroborated this. Maybe recessions are inherently unpredictable because they arise from unknowable network effects.

Thirdly, a highly indebted (pdf) world increases the danger of a financial crisis. Someone, somewhere has leveraged exposure to commodities and commodity producers. If that someone is highly interconnected with other funds, there's a danger that their collapse will trigger others, via heightened counterparty or fire sale risk. The question is: will the collapse be like that of LTCM, which had big spillovers or like that of Amaranth, which didn't? (I stress that what matters is not so much the level of total debt as the distribution thereof .)

Fourthly, there's less chance than before that macroeconomic policy could stabilize economies in the event of recession or financial crisis. Conventional QE is less effective when bond yields are low. And governments' commitment to austerity makes it less likely that we'd get timely countercyclical fiscal policy.

Now, I don't say all this to offer reasons to expect shares to fall further. It's also a cliche because it's true that bull markets climb a wall of worry. If you want comfort, the fact that yield curves are upward-sloping points to us avoiding recession. Personally, I think most efforts to time the market beyond obeying the "sell in May, buy on Halloween" rule are mistaken.

Instead, my point is a simple one that's been made before. It's that stock markets' falls remind us that high leverage, financial interconnectedness and the zero bound all combine to make us vulnerable to recession or crisis.