Would it be heresy to say employment numbers are rubbish for forecasting future space demand? The growth in employees in a given location should lead to floorspace demand, shouldn’t it? Jobs = floorspace need = leasing activity = rental growth. Easy.

But where do the employment numbers come from? They rely on government data; namely, the Annual Business Inquiry, which in turn relies on employers to provide the details of who works for them and where. But reported location could, because of modern working practices, differ considerably from actual location.

Take our team as an example: in EMEA Research based in Canary Wharf, London we have about 30 people. On any given day there is somewhere between 5 and 25 people physically present, meaning that the others are working from home or from offices elsewhere in London or Europe… The implication is true physical occupancy of quite a bit less than registered occupancy. Gross this up across full tenancies and across markets and you could be talking serious numbers…

So what? Playing devil’s advocate to our own forecasts (and everyone else’s) could we end up being disappointed by the real estate implications of employment growth when it comes? Could employment growth in one geography mean physical demand for space in another? Forecasts are meant to be an indication of trend, not an exact predictor of take-up volumes, so maybe it’s not the end of the world. And it underlines the importance of understanding the supply side when forecasting – this being a real estate agency’s USP. But changes in workplace practice will demand a rethink of forecasting methods, and potentially a radical one. Where is there room in econometrics for workplace strategy?