Join our community of smart investors

The shipping forecast

Shipping costs have collapsed. History, however, tells us that this is no reason for equity investors to worry
February 13, 2020

Shipping costs have slumped. The Baltic dry index (BDI) has fallen by 80 per cent since August. In theory, this could be a recessionary indicator as it signifies reduced demand for freight transport. And there is a worrying precedent. The BDI collapsed by over 90 per cent in 2008, which – we now know – warned us that 2009 would see a slump in output around the world.

Should we therefore be worrying now? Probably not. For one thing, the link between the BDI and economic conditions has not always been strong. Since 1997, the correlation between it and US capacity utilisation (a measure of the state of the global economy) has been only 0.12. That’s very weak. One reason for this is that BDI tells us not just about the demand for ships but the supply of them. One or two ships being newly built, or coming out of dock after maintenance, can send the index plummeting.

This said, there has been a more significant correlation between the BDI and both gilt yields and the break-even inflation rate, the gap between conventional and index-linked yields. A higher BDI is associated with higher levels of both. Bond markets seem to think that higher shipping costs – whether because of higher demand for ships or reduced supply of them – lead to higher inflation. The fact that gilt yields and inflation expectations have fallen recently while the BDI has slumped thus conforms to the historic pattern.

 

 

So much for the BDI as an indicator of current conditions. But what about its lead indicator properties, what does it predict? Not much. Yes, since there has been a correlation recently between the BDI and changes in US capacity utilisation in the following 12 months, with a low BDI leading to falls in utilisation – that is, weaker economic conditions. But this correlation has only held since the financial crisis. Before then, the BDI had no predictive power. This suggests that while there is a reason to fear that the BDI points to worsening economic conditions, it is not an especially strong one.

What’s more, history tells us that a weak BDI does not predict falling share prices. Since 2010, there has been no correlation between it and subsequent annual returns on the All-Share index. And before 2010, the correlation was actually negative, meaning that a low BDI predicted better returns. For example, the last time the BDI was lower than it is now was in January 2016. Had you bought UK equities then you would have seen a 15 per cent price rise in the following 12 months. And the low point for the BDI in the financial crisis came in November 2008. Although equities fell in the following five months they recovered strongly thereafter and by November 2009 they were well up from a year previously.

A low BDI, therefore, is no reason to sell equities (but no reason to buy, either.)

One reason why the BDI tells us so little about future returns is simply that it is so volatile, which means it contains a huge ratio of noise to signal. Since 1997, its (arithmetic) average annual change has been almost 20 per cent. But the standard deviation around this has been a massive 73 percentage points. This implies that we’d expect to see 50 per cent drops in it around one-sixth of the time. But, of course, we don’t see slumps in share prices or output that often. Paul Samuelson, the founder of modern economics, used to say that the stock market has predicted nine of the past five recessions. But the BDI has predicted even more.

Another reason why the BDI doesn’t predict equity returns is simply that stock markets are not wholly inefficient. They do discount some things. Insofar as the BDI contains any information about future economic conditions, investors price it into shares quite quickly.

Given its volatility, however, it’s unclear whether the BDI tells us anything about the economy that we couldn’t work out anyway. In 2008, for example, the strongest clue that the economy was heading for trouble was not that the BDI was falling – it had done so in 1998 and 2006 without disaster striking – but that the banks were collapsing. Similarly, the main reason to worry now is not that the BDI is low – it was even lower in 2016 without any ill-effect – but that the coronavirus might have a long-lasting rather than merely temporary effect on China’s economy.

There are very few reliable lead indicators of equity returns. The BDI is not one of them.