17/9/20: Exploding errors: COVID19 and VUCA world of economic growth forecasts

 

Just as I covered the latest changes in Eurozone growth indicators (https://trueeconomics.blogspot.com/2020/09/17920-eurocoin-leading-growth-indicator.html), it is worth noting the absolutely massive explosion in forecast errors triggered by the VUCA environment around COVID19 pandemic.

My past and current students know that I am a big fan of looking at risk analysis frameworks from the point of view of their incompleteness, as they exclude environments of deeper uncertainty, complexity and ambiguity in which we live in the real world. Well, here is a good illustration:

You can see an absolute explosion in the error term for growth forecasts vs actual outrun in the three quarters of 2020 so far. The errors are off-the-scale compared to what we witnessed in prior recessions/crises. 

This highlights the fact that during periods of elevated deeper uncertainty, any and all forecasting models run into the technical problem of risk (probabilities and impact assessments) not being representative of the true underlying environment with which we are forced to work.  

17/9/20: Eurocoin Leading Growth Indicator 3Q 2020

 

Eurocoin, CEPR & Banca d'Italia leading growth indicator for Euro Area economy is pointing to renewed weaknesses in the Eurozone economy in August, falling to its lowest levels in the COVID19 pandemic period:

As the chart above shows, Eurocoin fell from -0.5 in July to -0.64 in August, its lowest reading since June 2009. The forecast September indicator is at -0.30. Through August, we now have five consecutive months of sub-zero readings. Based on July-August data and September forecast, we are looking at a GDP contraction of 3.5 percentage points in 3Q 2020. This is mapped out in the chart below:

As the chart above shows, average annual growth rate in the Eurozone for 2020 is now sitting at -6,33 percent, far worse than the previous low of -0.575 in 2009. In quarterly readings, we now have two actual and one forecast quarters of 2020 all performing worse than the peak of the Global Financial Crisis / Great Recession contraction (see green entry in the chart above).

As before the COVID19 crisis, Eeurozone economy is performing woefully. On no time horizon did Euro area manage to achieve average annual growth of 2% (chart above).

17/9/20: Stonks are Getting Balmier than in the Dot.Com Heat

Via Liz Ann Sonders @LizAnnSonders of Charles Schwab & Co., Inc. a neat chart summarizing the madness of the King Market these days:

Yeah, right: PE ratio is heading for dot.com madness levels, PEG ratio (price earnings to growth ratio or growth-adjusted PE ratio) is now vastly above the dot.com era peak, and EPS is closer to the Global Financial Crisis era lows. 

What can possibly go wrong, Robinhooders, when a mafia don gifts you some chips to wager at his casino?

13/9/20: COVID19 Update: Nordics

 I have not updated the controversial comparatives between the Nordics (ex-Sweden) and Sweden in terms of COVID19 pandemic figures for some time now. For those of you who are out of the loop in this controversy, 

• Sweden imposed weak restrictions (basically none) on mobility and work environments in the wake of the pandemic, pursuing the strategy of 'herd immunity';
• Other Nordics imposed severe crack downs on mobility and social and work environments in response to COVID19 pandemic.

There has been a lot of controversy as to the effectiveness of the lack thereof of the Swedish strategy.

So here are the key data trends and points:

Key takeaways from the above:

• As of September 12, Sweden has the total number of 89,377 cases. Nordics ex-Sweden, normalized to the Swedish population size had 35,277 cases, or 39.5% of the total number of Swedish cases. So to-date, Sweden has experienced significantly greater exposure to the virus than its Nordic countries counterparts.
• As of September 12, Sweden has the total number of 5,829 deaths. Nordics ex-Sweden, normalized to the Swedish population size had 2,188 cases, or 37.5% of the total number of Swedish deaths. Sweden also experienced much higher death counts from the pandemic than its Nordic countries counterparts.
• Notably, per first chart above, since 26-27th of August, Nordic ex-Sweden have experienced a rather substantial re-acceleration in new cases, in line with the rest of the EU27 (see more on this here: https://trueeconomics.blogspot.com/2020/09/12920-covid19-update-us-vs-eu27.html). Sweden is only appearing to show signs of such re-acceleration starting with the end of the first week of September.
• The above change in trends for new cases is yet to translate into change in trends in deaths: in the last 7 days, average daily new deaths for Nordics ex-Sweden, adjusting for population differences, was 0.83. The same number fo Sweden was 1.57. In the prior 14 days, the averages were reversed at 1.034 and 0.571, respectively.

Watch the new wave developing in days ahead... 

13/9/20: COVID19 Update: Russia and BRIICS

 Updating my long-overdue charts for Coronavirus pandemic in Russia:

One worrying trend is the uptick in smoothed data for new cases in the last two weeks and a smaller corresponding increase in new deaths, as noted in the chart above.

Looking at the controversial mortality rates:

To highlight the above, consider the following differences between some key countries/country groups:

Note: China does not enter the comparatives, since it has, officially, less than 100,000 cases to-date.

12/9/20: COVID19 Update: Worldwide Cases and Deaths

The global COVID19 pandemic is refusing to go away. Day 247 and the dynamics of new cases and daily deaths remain as worrying as ever, just as the public and the politicians have largely settled into the strange 'ignore and forget' pattern of thinking about the threats.

Here are the daily new cases and deaths charts:

Key takeaways from the above two charts are:

1. There is no amelioration in the pandemic dynamics in terms of new cases: we are still at the peak levels of contagion worldwide.
2. Earlier hopes for decline in new cases worldwide has now been replaced by a re-acceleration in the new cases arrivals.
3. Historical average daily new cases is 115,366. Last 30 days average is 263,940 new cases per day and last 7 days average is 267,832. The pandemic is getting worse, not better.
4. Historical average daily deaths is 3,710. In the last 30 days, the average has been 5,603 and in the last seven days, 5,863. 
5. The peak of daily deaths occurred back in mid-April owing to a number of factors, that can be considered as exceptional. These include heavy concentration of COVID19 cases, lack of medical experience in treating COVID19 cases, demographic biases in early wave of COVID19 cases, and so on. On-trend, the peak was around 7,000 daily deaths.
6. Since the peak, daily deaths counts did not decay beyond the local trough around the end of May - a trough that was less than 50 percent lower than the peak (not an impressive moderation for a pandemic). The trough was set at around 4,050 daily deaths.
7. Since then, however, matters worsened once again, with a new local peak reached at around 6,000 cases in the last week of July. Deaths counts have been bouncing around 5,600-5,700 daily cases since then.

These worrying numbers are confirmed by the dynamics - the rates of change in the series:

As you can see from the chart above, the data is exhibiting a risk of rapid re-acceleration in the pandemic dynamics in terms of both, new case numbers and deaths. Of course, the data is volatile, so it is hard to tell if the new re-acceleration of growth rates to 15.8 and 24.4 percent for new cases and deaths, respectively, is going to hold. Nonetheless, nominally, we are no longer witnessing the second derivative (acceleration) moderating. Which means we are at a risk of rising severity of the already horrific trend. 

Here is a handy summary table of comparative growth rates:

Please note, these are growth rates that are already smoothed by using 7 days average! 

Fir numbers geeks amongst us, here is a summary table of comparatives for 33 countries, plus the EU27, that have more than 100,000 cases reported. Last time I updated these figures on August 30, there were 30 countries in this club.

Due to deterioration in the global pandemic conditions, EU27 is now ranked as an 'average' (statistically) territory. The U.S. remains 'worse than average'. 

The same table by aggregating countries by some internationally-recognized groupings:

And a descriptive statistics table for all countries with > 100,000 cases:

In summary, the pandemic is not going away. We are months before the start of the 'normal' flu season in the Northern Hemisphere, when temperatures outside force much of the activity - work, social, personal - indoors and into closer proximity to others. More inclement weather is coming to the largest concentrations of people on Earth, not only in Europe (already living through the second wave of the pandemic) and North America (still struggling with the first wave), but also to Asia-Pacific and the likes of India (which already outpaced Brazil to the second place in terms of infections after the World Series leader, the US of A). 

The brighter news on the horizon, currently, are the news relating to the vaccines development. But these vaccines are unlikely to hit mass markets any time soon, at least not before the October-November flu season onset. 

12/9/20: COVID19 Update: U.S. vs EU27

After some pause, the latest update for the U.S. vs EU27 Covid19 pandemic comparatives: 

• In summary: the basket case of the U.S. is getting somewhat challenged by the second wave sweeping across Europe, and we are still months away from the traditional flu season.

So here are the details:

As it says on the 'tin': new cases continue to climb in the U.S., albeit at slower pace of daily new additions than before. EU is showing acceleration in new cases growth, but for now, the rate of new cases additions in Europe is still well-behind that of the U.S. 

In terms of deaths per capita, the U.S. continues to pull away from Europe:

Key takeaways from the above graph are:

• The U.S. has a vastly higher death rate per 1 million population than the EU27 rate: current death rate per 1 million of population in the U.S. is 590.0 against the EU27's 318.3. Ouch!
• Put differently, current U.S. death rate per capita is 85 percent above that for the EU27. Going for the double within the next month, based on the trend.
• This means that overall counts of deaths in the U.S. are now above the EU27, since July 12. And things are getting worse: current excess gap is at +52,015 up from +29,428 a month ago and +1,841 two months ago.
• Adjusted for population differences, the U.S. has 88,868 more deaths (call them the excess deaths) than the EU27. Adjusted for the later onset of the pandemic and population differences, America's death toll from COVID19 is to-date is 98,786 higher than that of the EU27. Time to blame Obama and China, or, if that fails, Putin.

Daily new case counts:

Daily new deaths counts:

Key takeaways from the last two charts:

• Europe has now entered a full-fledged second wave of contagion.
• This contagion is more widely distributed across countries, which makes it less 'visible' compared to the first wave. 
• Numbers counts in Europe are now at the prior wave's peaks, and massively above the levels that triggered original shutdowns and quarantines. 
• All of the above means that the deaths counts are rising, albeit off low levels, and will continue to rise.
• Meanwhile, in the (misnamed) Land of the Buffalo (yeah, Bison ain't Buffalo, folks) things are not getting better. They are just getting worse slower. 
• U.S. deaths are running at daily rates well-above the first cycle trough, although it is too early to call this a 'second wave', yet. New cases remain very high, albeit significantly off the first wave peak.

A handy comparative table:

Meanwhile, the stock markets are roaring like a drunken Boris in a sauna around hour 2 of steaming.

12/9/20: Deep Impact of COVID19 on economic sectors supply side

 

An interesting research from McKinsey on a range of COVID19 pandemic impacts across a broader range of industries -- from traditional to advanced (advanced industries group includes automotive, aerospace, advanced electronics, and machinery players):

Source: https://www.mckinsey.com/industries/advanced-electronics/our-insights/organizing-for-speed-in-advanced-industries. 

Notably, intangible areas of impact appear to have been largely positive for all sectors, but tangible, productivity-related areas show deterioration. Worryingly, there is a significant - by sector and by sample average - adverse shock to productivity and volatility of business outcomes. 

9/8/20: Ireland PMIs and Economic Activity Dynamics for August

Ireland's PMIs are signalling a cautious recovery in the growth dynamics across three sectors, with growth still underperforming historical averages.

Irish Services Sector PMI rose to a respectable 52.4 in August from 51.9 in July, with the latest index reading sitting 38.5 points above April 2020 COVID-19 pandemic lows. However, statistically, the index remains below historical average of 55.0 and the median of 56.8. In other words, second month post-contraction phase of the pandemic, Irish services sectors are still struggling to restore growth (not levels) in activity consistent with a robust recovery.

Irish Manufacturing Sector PMI fell to 52.3 in August from July's 57.3 reading. The series are generally more subdued than Services PMI, which means that August reading is statistically indistinguishable from the historical average of 51.5 and is bang-on the median of 52.31. Manufacturing activity swung 16.3 points between COVID19 trough and August reading. Overall, Manufacturing growth seems to have fallen off the post-COVID19 high.

Irish official Composite (two sectors) PMI is currently at 54.0 which is statistically at the historically median rate of growth. The series are too short to talk about averages and historical comparatives in any serious terms. 

Irish Construction Sector PMI (not included in the official Composite PMI) came in at 52.3 in August, up from 51.9 in July and 48.7 points above the COVID19 trough in April. Current reading is statistically above the historical average, but identical to the historical median. This suggests that much of the rebound can be down to seasonal and cyclical volatility, as opposed to thee genuine recovery. 

Here is a summary chart of the three sectors dynamics:

I compute my own GVA-shares-weighted 3-sectors Activity Index, using all three sectoral PMIs reported by IHS Markit. The 3-Sectors Activity Index currently sits at 52.4, down from 54.1 in July and up 30.1 points on COVID19 trough. The current growth in economic activity in Ireland is statistically below historical average, and historical median. And it has moderated from July high, suggesting that the economy is still struggling to recover levels of activity lost to the COVID19 pandemic.

8/92020: BRIC: Composite economic activity indicators

 Based on Markit's Composite PMIs, here are the BRIC economies composite economic activity indicators for 3Q 2020 to-date (July-August). 

Please, note: Manufacturing PMIs for BRIC economies were covered here: https://trueeconomics.blogspot.com/2020/09/8920-bric-manufacturing-pmis.html, and Services PMIs were covered here: https://trueeconomics.blogspot.com/2020/09/8920-bric-services-pmis.html. 

• Brazil Composite PMI for 3Q 2020 currently sits at 50.6, barely above the zero-growth line of 50.0 in statistical terms. This represents a major improvement in growth momentum compared to an outright depressionary reading of 31.8 in 2Q 2020 and a swing of 18.8 points - a respectable reversal of momentum, although not a signal of an appreciable recovery from the recession.
• Russia Composite PMI for 3Q 2020 is at blistering 57.1, suggesting a genuine recovery, albeit one-sided, driven by services sector rebound. COVID19 pandemic low was recorded in 2Q 2020 at abysmal 32.6, implying latest swing from the trough of massive 24.5 points. This does appear to be consistent with a robust recovery momentum, albeit with some caveats. This is the highest reading for any quarter since 3Q 2006 and the third highest reading on record.
• China Composite PMI is at 54.8 so far through 3Q 2020, an improvement on growth-signalling 52.6 reading in 2Q 2020 and up respectable 12.8 points on recession trough in 1Q 2020.
• India Composite PMI managed to rocket to a still-recessionary 41.6 in 3Q 2020 to-date, up on the recession trough of 19.9 in 2Q 2020 - a swing of 21.7 points. Still, the economy remains in a pronounced recession and 3Q 2020 so far is showing signs of exacerbated contraction building on 2Q 2020 collapse in activity.

For BRIC economies as a whole, the chart next shows GDP-weighted and GDP-shares weighted BRIC Indices of activity, compared to Global Composite PMIs:

Overall, BRIC economies growth momentum is still subdued and largely performing worse than the Global Composite PMI indicator.

8/9/20: BRIC: Services PMIs

Services sector activity as reflected by PMIs from the BRIC economies is now available for August, so here are the top numbers: 

In terms of actual readings, and do recall, quarterly PMIs referenced above are averages over three months period, so 3Q 2020 data is only covering July-August 2020.

• Brazil Services PMI was nowhere near the insane reading posted by the country Manufacturing PMI (see post here: https://trueeconomics.blogspot.com/2020/09/8920-bric-manufacturing-pmis.html). Services index came in at 46.0 for the period of July-August (3Q 2020 to-date), up on disastrously low 30.3 in 2Q 2020, but still well below 50.0 line of zero growth. Reading PMIs, this means that the sector activity continued to contract in 3Q 2020 so far, on top of the already sharp contraction experienced in 1Q 2020 and 2Q 2020.
• Having set no records in Manufacturing, Russia Services PMI came out with a massive and seriously surprising print to the upside in August. As the result, 3Q 2020 to-date Services PMI rose to 58.4 for the highest reading since 2Q 2009. As massive as the print is, it is pretty 'normal' for volatile Russian services data. Still, the recovery it signals is sharp, as 2Q 2020 COVID19 trough was at a misery-inducing 32.0. The implied trough-to-peak swing is jaw-dropping 26.4 points.
• China Services PMI rose in the first two months of 3Q 2020 to 54.1 from already expansionary 52.6 in 2Q 2020. Trough-to-peak COVID19 swing is now at 13.7 points, and the latest reading is the highest since 4Q 2010, when the index stood at 54.2.
• India Services sectors are still in sharp contraction. Recall: in 2Q 2020, India Services PMI crashed, smashed, collapsed, melted down, or whatever else you might call, falling from 54.1 in 1Q 2020 to 17.2 in 2Q 2020, the lowest reading for any BRIC economy in any sector at any time. So far, in 3Q 2020, the index is running at 38.0, which implies that India's services sectors continue to contract from already reduced activity in prior quarter. In the light of this super-sharp recessionary dynamic, it is impossible to reconcile Manufacturing sector PMI and Services sector PMI in this economy.

Overall, BRIC Services Activity Index - a measure I compute using a range of data inputs, including Markit's PMIs - came in 49.9 in 3Q 2020 (to-date), an improvement on 2Q 2020 reading of 40.4 and above 1Q reading of 44.9. Nonetheless, across the four largest EM economies, Services activity continues to contract for the third quarter in a row, nominally, and it is standing still statistically. In this, BRIC economies are distinct from the Global Services PMI indicator, which rose from 35.6 in 2Q 2020 to 51.3 in 3Q 2020 (to-date). 

Stay tuned for BRIC Composite PMIs next.

8/9/20: BRIC: Manufacturing PMIs

Updating BRIC Manufacturing PMIs - with some delay (sorry, start of the academic year):

 These are quarter averages, with July-August for 2020. And wee have:

• A mad print for Brazil. In the last four months, Brazil Manufacturing PMI went from 38.3 in May to 51.6 in June, 58.2 in July and a totally incredible 64.7 in August. As PMIs are indicators of month-on-month activity changes, and not comparable year-on-year, all we know is that there is a massive boom in the sector from the lows of the COVID19 pandemic recession. But we do not know if we are close to pre-COVID19 levels or close to the pre-COVID19 trends or anywhere, specifically. Still, if 64.7 reading in August is a genuine indicator of activity, we are seeing real recovery in the economy. In fact, July and August are now two highest PMI readings months in history of the series. On a quarterly average basis, we are at 61.5 so far for 3Q 2020 which is the highest in history, with the prior historical high registered in 1Q 2010 at 56.3. Current trough-to-peak swing in Manufacturing PMI for Brazil for the COVID19 period is incredible 19.5 points.
• Russia, in contrast, continues to show signs of weaknesses in the Manufacturing sector, with 3Q 2020 PMIs so far running at 49.8 - statistically reflecting zero growth. Notionally, this marks the sixth consecutive quarter of PMIs below 50 in nominal terms and a seventh consecutive quarter at or below 50. Current trough-to-peak swing in Manufacturing PMI for China for the COVID19 period is sharp at 10.8 points, and this before we establish any recovery (over 50.0) momentum.
• China Manufacturing PMI is currently averaging 53.0 for 3Q 2020, the highest reading since 4Q 2010. China posted statistically zero growth - PMI at 50.4 in 2Q 2020, on foot of a significant, but not catastrophic, contraction in 1Q 2020 at 47.2. Current trough-to-peak swing in Manufacturing PMI for China for the COVID19 period is relatively moderate at 5.8 points.
• India Manufacturing PMI for 3Q 2020 is at 49.0, having risen from the recession trough of 35.1 in 2Q 2020, and trough-to-peak swing is at 13.9 points.

Overall, BRIC Manufacturing activity index is at 52.3 as of the first two months of 3Q 2020, up from 45.0 in 2Q 2020 and 49.1 in 1Q 2020. The trend is for a substantial improvement over time, with the trough-to-peak swing currently at 7.3 points.The index is outperforming Global Manufacturing PMI that currently sits at 51.2 for 3Q 2020, up 7.6 points on the trough in 2Q 2020. 

Stay tuned for BRIC Services PMIs and Composite PMIs next.