There is no doubt about it: the UK and many other European countries are already in a depression. We think of a depression as even worse than a recession – as a sustained, long-term downturn in economic activity in one or more economies. A depression is generally deeper and longer lasting than a recession.

As the daily drama of the Covid-19 count continues to scare everyone senseless, there is another a key indicator of economic rather than physical health that we also need to pay close attention to. That is how much unemployment there actually is, how much it has increased already and how much it will increase in the not too distant future. As everyone has hunkered down, many workers are not at their jobs and unemployment is growing. The problem is we have no idea how much it has risen already and how big any increase will be. The levels of unemployment are likely to be beyond anything in living memory – as will be the speed of the increase.

Social distancing has had a catastrophic effect on otherwise viable businesses. Many firms are closing. With no clear vision of the length of the shutdown or the nature of the recovery, their future is uncertain. There has never been such a concentrated business collapse. The government has tried to respond but it has no idea of the scale of the problem it is going to have to deal with. We made some back of the envelope calculations and they are scary.

We already have some evidence of the extent of the collapse in business and consumer confidence, which fell as much between February and March 2020 as they did in the year from 2007-8 just before the start of the Great Recession. The collapse in activity and the accompanying rise in unemployment looks to be at least 10 times faster than in the Great Recession of 2007-9. The Cips UK purchasing manager’s index (PMI) for the period 12-27 March was published this morning and showed a record fall exceeding the previous record seen at the height of the financial crisis. Employment fell at the fastest rate since June 2009. Service PMIs for the eurozone were even worse.

The Great Depression is the benchmark for depressions. We only have annual data but that shows that between 1929 and 1933, the US unemployment rate rose from 3.2% to 24.9%. Over this four-year period, the number of unemployed Americans increased from 1.6 million to 12.8 million. In the UK, the unemployment rate rose from 7.2% to 15.4% between 1929 and 1932 and an extra 1.9 million people were put on the dole. By comparison, the Great Recession of 2008-10 was a minor affair, at least as far as unemployment was concerned. We have monthly data available which shows that the US unemployment rate rose by five percentage points between January 2008 and October 2009, while the UK rate increased from 5.2% at the start of 2008 to 8% at the beginning of 2010.

These increases in the unemployment rate cover periods of a year or more. The Covid-19 pandemic will relegate all previous records for short-term increases in unemployment into the third division.

We have a pretty good idea week to week in the United States of what is going on with unemployment. Every week each of the states reports how many unemployment insurance claims there have been and the US Department of Labor (DOL) publishes these estimates for the country. Nothing much appears to have happened in the US in the first two weeks of March. There were 211,000 new claims in the week ending 7 March and 282,000 in the week ending 14 March. But then the explosion came: for the week ending 21 March, 3.3 million workers filed for unemployment insurance; on Thursday 2 April the DOL reported that a further 6.6 million had filed new claims for the week ending 28 March.

The latest backward look at the US labour market for March was also published this morning by the Bureau of Labor Statistics that covered developments up to early March. It was truly awful, showing an unexpectedly large drop of non-farm payrolls of 700,000 and a fall of 3 million on another measure. There was also a rise in the unemployment rate from 3.5% to 4.4% in a month – the biggest monthly increase in 60 years. The data for April due to be released on 8 May will inevitably be even more appalling. This will only cover developments up to early March and is unlikely to pick up most of the increases that occurred in the second half of March, but April, May and June are going to look horrible.

Sadly, in the UK we have the least timely and least accurate labour market data in the advanced world. The rise in applications for universal credit in recent days looks devastatingly bad, but it only gives a rough guide to the scale of any rise in unemployment.

Every other country, including the US, will soon get data for March. The Office for National Statistics (ONS) published data on 21 March for the period November 2019 to January 2020 – a statistic that now has no value at all. On 21 April we will get data for December to February; in mid-May, January to March. It will not be until July that we will get an unemployment rate based solely on the economy under lockdown for March-May and by then it will be too late. Useless.

The Federal Reserve of St Louis has produced an estimate of the unemployment increase in the US based on the size of the occupations most of risk of being shut down by Covid-19. The estimate is staggeringly bad. Unemployment in the second quarter of 2020 in the US, they suggest, may top 32%. We have looked at estimates of the UK occupational structure using similar methods, by judging which occupations are most at risk from Covid-19, and assumed that in the short-run around 50% of the workers in these occupations will lose their jobs. Based on that, what is coming in the labour market looks horrendous.

We estimate that unemployment will rise by around 5 million workers from 1.34 million to over 6 million by the end of May, and it may well be much worse. That will take the unemployment rate to around 21% – more than five times its current rate of 3.9% – and quickly. Some of these workers will be temporarily furloughed and may return to work if and when there is a recovery, but this will only happen if their companies are solvent and have a market to sell into. This becomes a tougher call the longer the lockdown persists.

In short: unemployment is set to rise in the UK by a lot. And we should be tracking this rise on a daily basis, not six months after it has happened.

• David Blanchflower is a professor of economics at Dartmouth College, New Hampshire

• David Bell is professor of economics at Stirling University