Wake me up when November ends

We’ve got less than 48 hours until a second national lockdown in England begins, and who knows how long until it comes to an end. Pubs, hotels, and hairdressers’ doors will close again across England – as they have been in Wales since 23 October – with the significant tightening of restrictions representing the final dashing of hopes that this Autumn was one of fast recovery. Far from growing, our economy is now likely to be shrinking again – shrinking because of the return of the virus rather than the lockdown itself. This note considers the likely path of the economy and labour market in the months ahead, and assesses the renewed policy response.

The economic impact

When anticipating what the coming month has in store it’s hard not to be thinking of reliving the first lockdown and the economic pain it brought. But there are good reasons for not expecting the hit to economic activity to be as deep this time. Firstly, this is a very different lockdown: crucially, with schools remaining open we should avoid dragging as many parents out of the labour force. Secondly, fewer sectors will be closed entirely and there is more clarity about which those sectors are. Hospitality, leisure and non-essential retail are where the action is. The major, and semi-accidental, shutdowns that we saw of construction and manufacturing last time aren’t being repeated. Thirdly, many firms have now learnt to live with the virus, putting in place new ways of working. More restaurants are now ready to move to offering takeaways and construction sites are covid-secure.

But these are silver linings to a very grim cloud. The final reason why the economic hit now won’t be as big should be anything but reassuring: almost nothing could be. GDP fell by 25 per cent in the depth of the first lockdown, an historically unprecedented fall in GDP. While the economy has recovered since it was not at the pace we would have liked – hopes for a V-shaped recovery were always wildly optimistic. In August, with hospitality experiencing what in retrospect was more activity than we could sustain without seeing an increase in infections, the economy was still around 9 per cent down. So, even if the factors above mean that we don’t near the depths of economic pain seen in April, November is likely to see an economy operating at more than 10 per cent below what we were used to. A stylised monthly path of GDP is presented in Figure 1.

Figure 1: The economy is set to shrink again, but not by as much as in the Spring

So we are on course for a W shaped downturn. While the second dip should be nowhere near as deep as the first, we’re still talking about a second dip around twice as deep as that seen during the peak of the financial crisis (the peak-to-trough monthly GDP fall was around 7 per cent). Crucially, this hit reflects the reality of a resurgent virus, rather than the specific policy choice of the lockdown.

A stubbornly rising case load since the end of August will also have a material impact on the labour market. We do not currently have any official forecasts that reflect this reality, but the Bank of England will be the first to update its outlook this Thursday. We should expect its forecast for the unemployment peak to rise from the 7.5 per cent set out in August. While the latest official unemployment data only shows a rate of 4.5 per cent in the three months to August, our own (more timely) survey suggests it may have reached 7 per cent or 2.5 million unemployed in September.  Among 18-to-24-year olds we found a much bigger, 10 percentage point, increase in unemployment since February, implying one-in-five young people out of work (see Figure 2). This risks a lasting impact on those young people’s labour market outcomes – including how many of them remain attached to the labour market. While half of those previously furloughed were back at work in September, one-in-ten no longer had a job. That rate of exiting furlough for unemployment was one-in-five for young workers, as well as BAME workers.

Figure 2: Employment effects vary widely across groups

The policy response

As with the first lockdown, the policy response will have a big impact on what this lockdown means for that rise in unemployment. The welcome return of a full, economy-wide, furlough scheme announced by the Prime Minister on Saturday night will slow redundancies significantly during the month of its duration – and not just those in the sectors now being locked down again. The policy of paying 80 per cent of full or partially furloughed workers wages for hours not worked, while only asking the employer to pay the National Insurance and pension bills, is a return to the Job Retention Scheme exactly as it existed in August.

But many employers, now recognising that this is anything but a short sharp crisis have already set redundancies in train. And more importantly we need to understand that the majority of this rise in unemployment is being driven by huge falls in the rates of unemployed people finding new jobs rather than in faster rates of job losses. Even in September vacancies remained a third lower than normal, so unemployment is likely to continue rising even if the renewed furlough scheme slows the pace of redundancies.

How many people are likely to end up furloughed this time is hard to say with any certainty. A reasonable estimate would be 4.7 million, based on the numbers still furloughed in non-shut down sectors according to the most recent surveys (1.1 million in early October) plus the number of workers furloughed in hospitality, leisure, retail and other services (e.g. health and beauty) during the last lockdown scaled down to reflect lower employment in those sectors (see Figure 3 below). With this level of take-up, the gross spending on this scheme would be £5.3 billion in November.

Figure 3: Millions of employees are set to be furloughed again

Support for the self-employed is also being ramped back up for this second lockdown. The Self-Employment Income Support Scheme (SEISS) will continue and the Chancellor has increased its short-term generosity, with grants worth 80 per cent of previous earnings in November (up from 40 per cent that will continue to apply in December and January). As a result, the maximum payment over those three months will now be £5,160.

This additional support is welcome given that the scale of loss of work for the self-employed has been much higher than for employees. Their return to work has been slower too, with around one in seven still having no work this Autumn. But continuing largely with the scheme as it existed for the initial lockdown means we will see a repeat of its pretty awful targeting. The original SEISS payments gave £1.3bn to people who had no real income fall at all. While the scheme’s guidance does now make clear the need to have been affected by the pandemic, many will again receive large payments despite being largely unaffected, while over half a million workers who received no support despite being without work entirely so far will again miss out because of eligibility rules. With this crisis having gone on much longer than initially hoped, we should have by now moved to a system of support for the self-employed that is more generous in terms of eligibility but ties payments much more closely to actual income losses faced.

The benefits system will also be playing a bigger part during this lockdown than the last as fewer people rely on furlough and more are unemployed. The significant increases the Government put in place during the first lockdown largely remain in place, including the just announced extension to the suspension of the Minimum Income Floor to the end of April. This suspension is ensuring that the self-employed who may not qualify for SEISS are able to receive more support via Universal Credit when their earnings are very low. But with higher unemployment set to be with us well into 2021 (and probably for years to come) it is time the Government made clear that it will not be pressing ahead with plans to cut £1,000 a year from 6 million households next April – plans that would reduce the incomes of the bottom fifth of the population by 7 per cent overnight. The impact of this across the distribution is shown in Figure 4. Delaying the (in my view inevitable) announcement on this is merely creating huge uncertainty for the lowest income families in the country.

Figure 4: The cut in benefits next year would see the poorest households lose 7 per cent of their income

On firms, the Chancellor has also set out plans to largely repeat the approach followed in the first lockdown. Direct grants (Local Restrictions Support Grants) will be focused on the shutdown sectors, with firms entitled to a flat £3,000 per four weeks payment, going to over 600,000 business premises. This will leave gaps for the supply chains (think of the laundry business that won’t get this help, but will be without work because they only supply now shut down hotels), although Local Authorities in England are also getting £1.1 billion to support businesses more generally as they see fit. The Chancellor is also making it easier for firms to access 100 per cent government guaranteed Bounce Back loans by extending the deadline for applying for them to the end of January, and allowing business that have previously taken out such a loan to top it up (if they haven’t already borrowed more than either cap of £50,000 or 25 per cent of previous turnover). In reality, the effectiveness of further loans will diminish as this crisis goes on, with some firms having already taken on as much debt as they can reasonably expect to be able to pay back. With very high default rates (the OBR estimates this could be as high as 40 per cent, which translates into write-offs in excess of £12 billion) expected on Bounce Back loans we are in reality now operating a system that is a messy hybrid of grants and loans, with little sense of what the end game is.

To complete the sense that this is a repeat of the policy response from lockdown I, Thursday is also likely to see the Bank of England upping its purchases of government gilts via Quantitative Easing from the £150 billion they have committed to so far. While this is likely to have little direct simulative effect on the economy or to significantly reduce interest rates in the short term, it will continue monetary policy’s unusual primary role in this crisis: ensuring that badly needed fiscal policy response continues to be delivered. While debates will continue about whether further marginal stimulus could be achieved via negative interest rates, the underlying reality is that fiscal policy is the primary tool for supporting the economy in this crisis (as Gita Gopinath of the IMF has set out). Borrowing this year looks set to be a peacetime record on every measure of around £400 billion, with the Bank of England providing a crucial supporting role in ensuring that – and the support for the economy it delivers – can take place smoothly.

Overall then what we see from policy makers is a broadly similar package to last time, so it is not surprising that its strengths and weaknesses are similar. On the strength side is the scale of support, which is very significant once again. However old problems remain, with the targeting of self-employment support standing out. New ones will also grow given the very real impact of the previous lockdown, with an approach relying on ever more loans for firms becoming less useful (or at least honest in expecting full repayment) over time.

An exit strategy?

While this provides a static analysis of the renewed policy response, we should dwell on the need to learn broader lessons – about the duration of, and exit strategy, from this lockdown.

One big distinction this time around is that the policy package is more optimistically time limited to match the intention to lift the lockdown in early December. Of course, whether in reality it is feasible to lift this lockdown then remains to be seen. And here we should be clearly learning the lesson of the crisis so far in two respects. Firstly, that significant uncertainty exists about the future path of the virus and the measures that will be needed to combat it – so it is better to be clear that plans for the withdrawal of policy support are conditional on that path rather than claiming to have a fixed plan. The insistence on pushing ahead with the latter is why we had the ludicrous situation last weekend of guidance for the Job Support Scheme (itself only announced on 23rd September and already heavily amended twice) being issued on the Friday night before it was due to replace the Job Retention Scheme on Sunday – only for the original Retention Scheme to then be extended on the intervening Saturday. Being clear on that conditionality is also going to be required now to avoid a total mess of which support is available to which part of the United Kingdom as lockdown timetables differ.

Those timetable for successfully lifting at least elements of this lockdown (it seems likely we’ll see non-essential retail restrictions lifted ahead of anything for hospitality) depend on big improvements in the wider systems for controlling the spread of the virus. Crucially, lockdown needs to be accompanied with sorting out NHS Test and Trace. Others will have more to say than us on what that means for the operational side of testing and app development. But for any of these systems to work we need much higher compliance with the regime than we presently see.

If we want people to do their bit in controlling the spread of the virus, even if they don’t feel ill themselves, then we need a step change in our approach to sick pay to ensure people feel they do not pay a high personal price for contributing towards our collective health response. Our current system comes in two highly flawed forms. Statutory sick pay provides an inadequate level of protection (£95.85) and excludes two million of the lowest earners. It needs to be raised and made available to lower earners. Instead of doing that the government has sought to address this problem with a much more generous £500 Test and Trace Support Payment. But because to qualify for this payment you need to be receiving benefits like Universal Credit (or be able to show that you face real hardship) seven in eight workers will not qualify for it. The result is a mess of a system providing too low or no support at all for people if we ask them to stay home.

Sorting out this sick pay mess isn’t just about an exit strategy from lockdowns – it would help to make a success of the lockdown. Policy makers need to remember that just because most of them can work from home if needed, lots of workers will still be going to their workplaces in the months ahead – indeed in most cases will only get paid if they do so. 59 per cent of people were working in their workplaces rather than at home in September. We estimate that even with renewed lockdown this will only drop to around 45 per cent.

Conclusion

England is heading into a second lockdown this week. The return of the virus has made tighter restrictions inevitable, and that return also means that always overoptimistic hopes for a V-shaped recovery now face the reality of at best a W-shaped one. While the second dip in economic activity should be much smaller than the first, it will still be huge by the standards of any normal downturn. So it makes sense that the policy response is also once again very large indeed, with most elements of the package that existed during the first lockdown being reinstated.

What there is little evidence of is learning the lessons of what did – and did not – work last time. And it is crucial during this lockdown is that we put ourselves on a better footing for fighting this virus not just week by week, but for the months ahead. The signs of a vaccine and better treatments riding to the rescue to some degree next year are good – but that is no excuse for doing a poor job of managing this pandemic while we wait for the cavalry to arrive.