Macro Weekly – It takes courage to forecast a recession

by: Han de Jong

  • US ISMs gloomy, PMIs less so
  • Eurozone inflation stuck around 1%
  • Boris Johnson’s election campaign appears to have started
191004-Macro-Weekly-1.pdf (214 KB)

Economists are not very good at forecasting recessions. I really should rephrase that. Economists are very good at ‘forecasting’ recessions, but usually only when the downturn has actually already started. They (perhaps I should say: we) are not good at forecasting recessions sufficiently early for policymakers, companies, investors and ordinary people to take mitigating actions. It takes courage to forecast a recession. Actually, some economists have plenty of this type of courage and they are more or less always forecasting things to get worse. They are the eternal doomsayers who enjoy their moments in the sun when they are occasionally and inevitably right. For some time after, they tend to be introduced as ‘Mr or Ms Such-and-Such, who correctly predicted this or that disaster’. As their predictions of the next disaster turn out incorrect they gradually disappear from the limelight.

US ISMs make gloomy reading

As I am looking at the continuous data flow from around the globe, I get a little nervous but also quite confused. The old reliable early indicators appear to be giving conflicting signals. The recent readings of the US ISM paint a worrying picture. The ISM business confidence gauge for the manufacturing sector fell to 47.8 in September, down from 49.1 in August where the consensus expectation among economists was that September would better than August. The September number was the second sub-50 reading. It suggests that the manufacturing sector in the US is under pressure, just as is the case elsewhere. The level of the ISM manufacturing is not consistent with an overall recession in the US, though. For that, a further decline to the 42 area would be required and that is far away.

The services sector has generally held up much better than manufacturing. That is probably not surprising as weakness in manufacturing is most likely predominantly caused by China’s slowdown and the uncertainty created by the trade war. Services are not or at least less affected by those factors. Having said that, many services are linked to the goods sector, so there is a question over how long the divergence between business confidence in manufacturing and services can last. That is why the release of the ISM’s non-manufacturing gauge also was a shock. That series dropped from (a strong) 56.4 in August to 52.6 in September. Quite a tumble!

Rival indices painting less pessimistic picture

So why am I confused, you may ask. Well, the Markit PMIs paint a different picture. They were released earlier and both US series improved in September. The ISM has a much longer history (it goes back to 1931 and the manufacturing series as we know it now goes back to 1948; the ISM services has a shorter history). As it has a proven track record, I am inclined to put more emphasis on it than on the Markit PMIs. It must be said though, that the advantage of the Markit PMI is that this survey is taken in over 30 countries on a consistent basis. The US wasn’t the only country where the Markit PMIs improved in September. The Chinese Caixin PMI’s improved as did, for example, the Canadian one, while the Brazilian PMI reached its highest level in 18 months.

As a bit of a challenge to the improved Chinese Caixin PMI, Korean exports to China were down 21.8% yoy in September, hardly a sign of a strengthening Chinese economy. Overall, Korean exports were down 11.7% yoy, a touch better than August’s -13.8%. China makes up around a quarter of Korean exports.

Another sign of weakness comes from Japan where the well-established Tankan survey shows a continued decline of business confidence. Like in other countries, non-manufacturing has held up better than manufacturing. However, the most recent reading isn’t anything to write home about.

The US non-farm payrolls for September were mixed. The economy added 136,000 jobs and there were revisions to the previous two months of +45,000 in total. Taken together, that was roughly in line with expectations, but it is clear that job has slowed. The manufacturing sector lost 2,000 jobs, the second negative monthly number this year. It is reasonable to think that the modest loss of manufacturing employment is a sign of weakening of the industrial sector. Nevertheless, the unemployment rate fell from 3.7% to 3.5%, the lowest unemployment rate in over 50 years. There was a time when economists thought that a tight labour market would trigger an acceleration of wage growth, but not anymore, apparently. Average hourly earnings were unchanged on the month while the year-on-year rate fell from 3.2% in August to 2.9% in September, the lowest in over a year. Wage inflation clearly isn’t a problem, creating further room for the Fed to ease policy. President Trump was quick to celebrate the drop of the unemployment rate to the lowest level in 50 years in a tweet. That overall jobs growth has slowed, that manufacturing actually shed jobs and that average hourly earnings growth slowed isn’t his fault, of course.

The global business cycle in short

So the picture is somewhat confusing, but my story about the global business cycle would be as follows. Chinese slowing, partly as a consequence of the policy of deleveraging is a key factor in the loss of growth momentum in the world at large. Yes the Chinese are taking measures to prop up growth, but so far not with decisive success. In addition, the trade conflict has had a hugely negative impact on global activity. Import levies are taxes and the rather random introduction of sometimes prohibitively high tariffs has negatively affected trade flows. Even worse, perhaps, is that the conflict and the tit-for-tat actions have created huge uncertainty. The economy does not like that.

While I believe that the Chinese slowdown and the trade conflict are the main culprits behind the global loss of momentum, they aren’t the only factors. Fed tightening in 2018 (through four rate hikes, but also through shortening of its balance sheet), the uncertainty over the Brexit process and the sector specific challenges in the (European/German) car industry also play a role and perhaps there is more.

Early this year, I was hopeful that most of these factors would turn around soon, but that really has been in vain. Despite stimulus, it looks to me as though the Chinese economy continues to lose altitude and the trade conflict goes through ups and down, hope and fear with uncertainty being the common denominator. True, the Fed has changed direction and the ECB has eased policy, but these measures take time to make themselves fully felt. For Brexit, see below. The problems in the European car industry appear more persistent than I had hoped.

So should we be forecasting recession?

The eurozone economy could certainly register a couple of negative numbers for GDP growth. But that may not be the important thing to watch. A significant rise in unemployment and bankruptcies would constitute a ‘proper’ downturn and to be honest, I don’t see that happening unless the US falls into recession. Such a development is not impossible, but, as we have said before, the US consumer is too strong to throw in the towel any time soon and that makes a US recession starting during the next couple of quarters unlikely. It is likely, however, that the US economy will continue to weaken in the period ahead. This process is led by corporate capex spending and will affect the labour market. But by the time this will all hit the consumer, the Fed’s easing will make itself sufficiently felt to prevent an actual recession. Fingers crossed. The twists and turns of the trade conflict, the presidential tweets, the impeachment inquiry as well as the approaching election campaign may all push the US business cycle either into a higher or an even lower gear.

Eurozone inflation stuck around 1%

Preliminary eurozone inflation numbers show that inflation remains stuck around 1%. Core inflation ticked up from 0.9% yoy to 1.0% in September while the headline series did the opposite. Given below-trend growth it is hard to see any meaningful acceleration of inflation any time soon. Following the Draghi logic it means that the ECB must, at some stage, do more to bring inflation closer to its target of ‘below but close to 2%’. But there is also ‘Klaas Knot’ logic following the recent interventions of the president of the Dutch central bank. One of his ideas is to redefine the inflation target and opt for a range. He has suggested 1-3%. I wonder what that would mean for monetary policy. The actual inflation numbers are right at the bottom of Knot’s range. I guess that would mean that somewhat, but not very loose policy is justified. Eurozone inflation has actually been remarkably stable in recent years. Core inflation has fluctuated in a range between 0.7% and 1.3% for five years or so.

Brexit saga continues

British PM Boris Johnson has so far failed to produce Brexit proposals to the rest of the EU that are acceptable in the eyes of the European parliament. I wonder if he intends to produce proposals that the EU can agree with and I also wonder if the European parliament will agree anything the UK government comes up with. And even if Johnson’s intentions are positive and constructive and the EU is equally constructive, how likely is it that a deal will be greeted positively by the UK parliament? Not very, I would think. This whole exercise is going to be fruitless I think, a waste of time and energy, but everybody must be seen to be going through the motions. If Johnson cannot expect to be successful, what is he playing at? Clearly, the campaign for the likely upcoming general elections has started. Johnson must convince the electorate that he has tried everything he possibly can to reach a deal and that the EU is completely unreasonable, all the more reason to want to leave.

We continue to think that elections are highly likely, but that the outcome is hugely difficult to predict. According to opinion polls, support for the Conservatives is edging higher and they have a 12 point lead over Labour. What is interesting is that support for the Lib Dems is rising. They were moving in the 6-10% range back in May, but have risen to 20% recently. That still leaves a large gap with the 36% of the Conservatives. As the election may effectively become another Brexit referendum, it is not impossible that the Lib Dems will continue to rise as they have a very clear position on the issue: they are against Brexit. That can’t be said about Labour. Therefore, Labour voters for whom trying to prevent (a hard) Brexit is important may flock towards the Lib Dems. This also creates a new perspective for Tory voters who do not like the idea of (a hard) Brexit. They might still support Boris Johnson even as he campaigns on a hard Brexit platform because they certainly do not like the idea of a Corbyn government. However, they may also move to the Lib Dems if such a vote lowers the hard Brexit risk while not putting Jeremy Corbyn in 10 Downing Street.

What is very unfortunate, it seems to me, is that the Lib Dem’s new leader, Jo Swinson, is relatively unknown. According to 10% of surveyed people approve of her, whereas 33% approve of Boris Johnson. The difference is explained by a large number of people not knowing Swinson: only 37% of surveyed people say they know her, against 98% in the case of the prime minister. The positive of that is that relatively few people say they disapprove of Swinson: 10% against 48% in Johnson’s case.