This week the US Portland Cement Association (PCA) revised down its forecast for the rise in cement consumption in 2016 to 2.7% from 4%. It also lowered its prediction for 2017, blaming political uncertainty around the presidential election, inflation and slower construction activity. Global Cement Magazine editorial director Robert McCaffrey pointed out on LinkedIn that he was surprised by the revision down in 2017 given the rhetoric by president-elect Donald Trump to invest in large infrastructure projects.

Clearly the PCA is playing it cautious as a politically unknown entity, Trump, slides from campaign trail promises to executive power delivery. Backing them up are the latest figures from the United States Geological Survey (USGS) that show that both cement production and shipments fell slightly in the third quarter of 2016. In the quarter before the election in November 2016 the cement market slowed down. The hard bit is working out why. As we pointed out in a review of the US cement industry in the May 2016 issue of Global Cement Magazine the PCA had previously downgraded its forecast in 2016 due to economic uncertainty despite strong fundamentals for the construction industry. Then, as now, the great hope for the US cement industry was infrastructure spending down the pipeline, at that time the Fixing America’s Surface Transportation Act. At this point it doesn’t seem to have had much of an effect.

Industrial and economic forecasters aren’t the only ones who have a hard time of it in 2016. Political pollsters have also been caught out. Surprises came from the UK’s decision to leave the European Union and the election of Trump. Neither result was widely expected in the media. As explained above, should Trump make good on his building plans then if any cement company based its plans on a forecast dependent on a Hilary Clinton win then it may have lost money.

The power of forecasts has even greater potential effects in developing markets where the corresponding financial risks and rewards are higher. After all, why would any cement company invest tens of millions of US dollars for a cement grinding plant or hundreds of millions for an integrated plant unless there was some whiff of a return on investment?

This then leads to the problems Dangote has reportedly been having with its plant in Tanzania. Amidst a flurry of local media speculation in late November 2016 about why its Mtwara plant had a temporary production shutdown, Dangote’s country chief clarified that it was due to technical problems. It then emerged this week that Dangote’s owner Aliko Dangote met with President John Magufuli to agree a gas supply agreement to the plant. The point here being that even if the market conditions and demographics seems conducive to profit, as is the case in Tanzania, if the local government changes any incentives agreed at the planning stage then everything can change. At this point forecasts based on data become moot.

There’s a great quote from the US pollster Nate Silver that goes, “The key to making a good forecast is not in limiting yourself to quantitative information.” In terms of election campaigns run at a time of upheaval that might mean listening to people more than looking at polling data. In terms of a cement company operating in Africa that might mean fostering links with the local government to ensure no sudden policy changes catch you off-guard. And in the US that might just mean cement company analysts have to follow Donald Trump’s Twitter account.