Is it sensible to use forecasts of trends in markets to meet future needs and requirements of the economy?

1

It is sensible to be guided by analysis of markets, whilst remembering that there is always a margin of error in predictions, and that unlikely or unexpected events are not impossible. As with cyclical analysis of the economy, it is also sensible to be aware of the projected length of forecasts, and that economic analysis is often limited by assumptions about rationality and marginality which economic actors might not share. A final caveat is that the ‘map paradox’ applies; no representation is perfectly accurate because, if it were, it would be the thing represented itself and not a representation.

Forecasts of trends help economic agents, governments, and monetary authorities make decisions about the allocation of resources and cash, and for example, central lending rates. In doing so, they might help eliminate waste, in the case of suppliers, which helps an economy move towards allocative efficiency.

Acting because of forecasts might also reduce or eliminate decision lag, in the case of governments and banks. Decision lag arises when authorities respond to a problem by taking time to develop a solution, create it, and then deploy it, by which time the original problem might have changed or disappeared. Governments also attempt, in making fiscal decisions, to secure the best value for the taxpayer whilst balancing costs against benefits. Forecasts which have been proven to be reliable will help in determining best value but will also reassure markets and taxpayers by offering a rational and transparent reason for actions. This was one of the reasons for the creation of the Office of Budget Responsibility in the UK in 2010.

Transport projections illustrate some of the problems which arise in using forecasts for planning. The creation of a solution to projected problems often creates other difficulties based on that solution. So, for instance, projecting a growth of cars and building motorways might lead to a phenomenon whereby people shift into much wider car ownership, congesting those motorways. This could be accompanied by economies of scale in the construction industry after an expansion of roads, allowing the building of new roads at low marginal cost. This could then lead to many more roads, much more car ownership, much more pollution, more accidents, much more dependence on credit to pay for cars, and higher costs for the freight industry on congested roads. Higher GDP would then result in more spending, and more demand for freight to be transported to shops and factories, leading to even more congested roads. If trends were to reverse because of a cyclical fifty- or twenty- year economic cycle located elsewhere (in technology or property markets) or if cars became too expensive or unfashionable, the whole model might collapse, leaving taxpayers with the bill.

Forecasts might also have not only a use in policy but as policy, because some data is market sensitive and affects the way in which decision makers operate. For example, publication of government forecasts on future population trends can affect market decisions about long-term growth, and therefore interest in government bonds. Similarly, calculations of profit and viability at different rates of interest and of different forms of investment might be seen in the pensions industry, leading to volatility and higher levels of risk and hedging.

Given that pensions based on a long period of low interest rates and moderate fiscal policy almost collapsed on the reception of a radical government budget in October 2022, nearly bringing down the UK economy, this is not a exaggerated worry. Bank of England projections of future inflation or interest rate changes, similarly, offer ‘forward guidance’ and influence market sentiment, which can change the forecast itself. If people think that higher inflation and therefore higher interest rates are imminent, they might moderate their behaviour in such a way that demand-pull inflation is greatly reduced.

Government forecasts have improved over decades, but policies which commit to infrastructure or housing provision based on long-term forecasts are subject to long-term failure because of changes in public taste, technology, or needs. This does not necessarily reflect ‘fault’ on the part of anyone. For instance, a great deal of the water and transport infrastructure of the City of Vienna in 1880-1914 was projected on a need to provision for a forecasted growth to twenty million people by the mid twentieth century. Consequently, by the twenty first century, when the population of Vienna was under two million, the sewage, water, and road systems were amongst the best and least pressurised in the world. London, by contrast, was forced into expensive construction of new infrastructure because it grew beyond all projection in the second half of the twentieth century and early twenty-first, leading to a crisis in the water and sewage system.

Incorrect or contingent forecasts may always be made and lead to such mistakes. Energy markets, in which closing North Sea oil and gas taps and running down storage because Russian gas was cheaper, are just one example where policy decisions not based on a combination of forecasts and limited interpretations of best value would have been more sensible. The 1975-6 IMF crisis in Britain, famously, started an era of austerity and the end of the post-war Keynesian consensus because of a forecast balance of payments crisis in the UK which, in fact, was based on what is now seen as flawed and mistaken evidence.

In private markets, dependence on ‘just in time’ global outsourcing in supply chains because of a forecast continuous growth in globalisation have given way to glocalisation and the return of national and regional manufacturing, which does nothing to reverse the great human cost in terms of deindustrialisation and structural unemployment of the intervening period.

Failed decisions based on the combination of forecasts plus a determination to have a policy which does ‘something’ and reverses something else—often promoted by consultants or policy entrepreneurs who are rent- or status-seeking, or who simply need attention—abound. It would be absurd, however, to write those sensible decisions based on evidenced projections, taken after a consideration and analysis of risk and countervailing trends, are always wrong. In fact, there is no serious substitute, and therefore there is a requirement for multiple forecasts which can be compared to each other and relied upon by reasonable policy makers. People should just recall constantly that every decision and every forecast is a risk, and that conditions and trends can and do change.