Think at London Business School
Esther Duflo and Abhijit Banerjee discuss potential ways forward, in conversation with LBS’s Elias Papaioannou
By Matt Munday
“The only function of economic forecasting is to make astrology look respectable,” the Canadian economist JK Galbraith once observed.
Andrew Scott, Professor of Economics at London Business School, shares some of Galbraith’s scepticism. At a moment like this, with so much uncertainty about the immediate and medium-term future, economic forecasts can only be made with a great deal of caution. Perhaps we are more in the territory of offering possible scenarios rather than precise forecasts, he suggests.
In his webinar discussing questions for leaders at this time of pandemic, Professor Scott set out his thoughts on what the economic consequences of the Covid-19 pandemic could be. For someone with a particular interest both in life expectancy and studying GDP it can seem like an especially gloomy and uncertain moment, he says.
For one thing, we cannot know when the current shutdown will end, although the median guess is perhaps the end of June. The scale of what we are living through is more or less unprecedented: this is a one-in-50 or even one-in-100-year pandemic event. This leaves us unprepared psychologically. And this is stressful, even without the fear of getting infected. “A trip to the shops makes you think about mortality,” he acknowledges.
The challenge for leaders, Professor Scott says, is to try to imagine a worst-case scenario while also working out what the best option might be, at the same time. This is hard to do.
A sharp and deep recession seems inevitable. There is some basic and harsh maths to be done. If the shutdown lasts for two months and this is followed by a complete recovery, there will still be a 3% loss in GDP. If the shutdown is for four months then the hit is 7%. You get to reach some very big numbers quite easily. The GDP figures for 2020 will inevitably be very bad.
In the UK, the Office for Budget Responsibility is predicting unemployment will soar by two million thanks to Covid-19. Similarly, job losses in the US will be extraordinary – many millions of jobs in all likelihood. This is why some economists are referring to this moment as “a great cessation”. It will not just be a normal recession.
It is an unusual downturn because it is self-inflicted, not the result of a financial crisis as in 2007-9. And because it is a different kind of economic slowdown the nature of the recovery may be different too.
Another unusual characteristic is that there is both a supply and demand shock taking place at the same time. People are stuck at home. There is much less spending going on. But supply is disrupted too. This is more complicated than a normal recession.
So what should we expect to happen? It is easier to make forecasts in more normal, steadier times. Unfortunately for economists, Professor Scott says, people are most interested in economics when we know the least.
It is possible to look at some comparable history. Recent research by Robert Barro et al looked at data from 43 countries after the great influenza 1918-20 (“Spanish flu”). Officially 39m people died, but perhaps many more. This was 2% of the global population. There was a 6% fall in GDP, and an 8% fall in consumption.
This took place at the end of World War One, when resources were already severely depleted. The world population was vastly smaller – around four times smaller. So can a comparison be made? The World Bank has also estimated that a pandemic such as Covid-19 might led to a 5% fall in GDP. The world is more interconnected today, and today’s shutdown more complete, and the private sector bigger. So the impact could be bigger.
Research by Richard Hughes at the Resolution Foundation has considered how quickly economies might recover. On average, after 10%-15% falls in GDP there could be recovery after three years. But few countries will display an average response – performance will vary greatly.
There four big unknowns which make precise forecasting impossible:
Different answers to each of those questions lead you to form different conclusions.
There is plenty of discussion currently about what sort of economic recovery there will be. What shape on the graph will growth over time look like: an L, a W, a U or a V?
The L shape is an unusual outcome, and barely a recovery at all – it is a flat line across the x axis. This would involve a sharp shutdown and permanent loss of output, as firms go bankrupt and job losses are not reversed.
The W – a kind of rollercoaster ride – depends on the path of the virus: will it come back in waves? The W involves a sharp shutdown, then mini recoveries and more slumps as either the shutdowns continue or waves of Covid-19 occur.
The U and the V patterns, a bit like the map of the London Underground, may mislead a little by their simple appearance. There will be variance in the figures as time passes, and growth will not be so smooth. The U shaped recovery involves a sharp shutdown, and a prolonged duration as the shutdown continues (or the shutdown has long-lasting effects), followed by recovery.
The V, the most optimistic letter of the alphabet in this context, represents a sharp shutdown and then rapid recovery as the shutdown controls are removed.
The speed of recovery will be asymmetric, that is, while the shutdown was fast, recovery will not be so fast. If it is V-shaped and the virus is managed successfully there will be one temporary shutdown. If it is a U, the length of shutdown will be longer and the recovery will be delayed. Lower demand will have amplification effects and further weaken recovery.
What to look out for next? The GDP figures will first be terrible, and then excellent. But we should look at the level (the size of the economy), not the growth rate. A 30% decline followed by a 20% increase still leaves the economy a long way back.
GDP data is a lag indicator. It tells you what has happened. The bad numbers we will see will be about the past. And of course initial GDP survey data always has errors in it. It is an aggregate figure. Individual stories are contained in it reflecting different ranges of performance. It may be new, not old firms which lead the recovery. Prices too may not all move in the same way. Some products and services may be highly sought after, others may become cheap commodities.
Economic policy is currently trying to flatten the curve, or shield the economy, just as the shutdown aims to flatten the mortality rate. The downturn will contain the seeds of its own recovery as people return to work and catch up on spending. But will the firms and jobs be there if shutdown triggers mass bankruptcies and unemployment? That must remain uncertain. Governments are using the “big bazooka” of massive fiscal intervention, in the belief that it is better to do too much than too little.
“The downturn will contain the seeds of its own recovery as people return to work and catch up on spending”
Talk of the scale of the economic challenges facing us all brings to mind the scale of disruption after a world war. There was a massive surge in government debt during World War TWo – around $2trn in US, or 10% of GDP.
As in war there will be a permanent loss of wealth as a consequence of this crisis. But wars last a lot longer than this war against the virus will. It is not like World War Two. The productive economy will not be skewed as much as it would be during a real war. But production is shifting, for example to manufacture ventilators. The private sector will not be as badly affected as in a real war. There remains a trade-off between public health and the economy.
As far as government debt is concerned, the deficit in the US will be larger than during the financial crisis, and perhaps as large as during a real war. In 2020 double-digit fiscal deficits are entirely appropriate. Governments should let debt and not the economy take the strain. This is a one-in-50-year pandemic, so you can take a long time to pay it back, Professor Scott says.
What are the risks inherent to the current policy mix? A key risk, according to Professor Scott, is that governments don’t know what they are doing. Can the government see who is most badly affected? The hardest hit may be hardest to help. How long will government stimulus continue? How will we pay it back?
The US and UK can probably cope. But there could be a big hit to the Euro system, with Spain and Italy challenged hardest of all. Inflation is less likely to be a problem, however. It is difficult to imagine demand roaring back so fast.
Economics is all about trade-offs, and we are currently engaged in one. Interestingly, the “Value of a Statistical Life (VSL), calculated in the US, is $10m. Every 10,000 lives saved is therefore worth $100bn. So even very large shutdowns are “worth it”, over and above any humanitarian concern, Professor Scott says.
“Even very large shutdowns are ‘worth it’, over and above any humanitarian concern”
It seems like a good moment to switch from an optimal strategy (what you would like to happen) to a robust strategy (which should work in most circumstances). As stated earlier, reliable forecasts are hard to make, but a robust forecast based on worst case scenarios can be achieved.
It is also time to stick to the mission of the business, not an outdated plan. Unfocused busyness is of limited use – “Farmers don’t sow seeds in winter,” Professor Scott says. Right now, being busy is not the same as being productive. We could hope for a “pitstop” recession: come back out and be faster.
Ultimately this crisis tests our faith in progress and science. It it is clear that there are some problems only government can deal with. The severity of the economic decline will be damaging, yet most people think this shutdown is the right thing to do.
The economy serves the people, and not the other way round. So maybe capitalism is not all about profits. There has been a realisation about what really matters to us. We are seeing a different kind of capitalism, not the end of it, Professor Scott adds.
The question is not if there is a role for government, but what is that role? Can we build climate change into our economic models? And can we prepare better for pandemics and for climate change?