Previous editions:  May 30, May 23, May 16, May 9, May 2, April 25, April 18, April 11, April 5

UGH. The first draft for today was "not very many new economic perspectives this week and nothing on coronavirus so here's how May ended." So much for that.

Today covering three main topics:

I hope you will also take a minute to read the stories on avoiding screwing up the next phase of the pandemic and also on scenarios about the future under the virus.

Thanks to those of you who called this week about assignments. My staff and I appreciate the work.

TAKING UNCERTAINTY ALL THE WAY TO ELEVEN. It's far too early to speculate on how riots will effect either the economy or the housing markets even in the cities most dramatically affected. It is a certainty that we will see more cases of coronavirus infection and more business closures either through riot damage or renewed infection control measures. What is uncertain now is the tolerance for enduring these measures - they are not being followed during the protests and this has not gone unnoticed by shutdown opponents. I am not sure of the implications of this, but it has dramatically increased uncertainty in dealing with the virus and it is not helping that public health authorities have shifted messaging on the virus yet again in response to protests.

The protest remembering George Floyd was estimated at 60,000 people. I dug into the history files, this would appear to be the largest protest ever in Houston. The city has largely been spared violent protests, at least over the last 100 years. Riots elsewhere have, however, changed our city in unexpected yet dramatic ways:

ABOUT THAT JOBS REPORT. Today's jobs report was outstanding. Most analysts (including me) expected 6-8 million additional job losses with an unemployment rate close to 20%. Instead employers added 2.5 million jobs in the best month since 1939 yielding an unemployment rate of 13.3%, down from over 16%. I'm going to give you a minute to let that sink in.

Now, for the rest of the story. The BLS report uses a macroeconomic model that is based on household surveys. People not working can be classified as "unemployed on temporary layoff" or they can be classified as "employed but absent from work." The BLS says people not working because of coronavirus should be classified as "unemployed on temporary layoff." I'm sure you can see where this is going, from the BLS:

If you read this and thought that the improvement in jobs in this report wasn't real but was instead a statistical artifact of a data collection error, that's OK. That's what I thought initially, too. But I don't believe the BLS would willingly release that kind of report. Instead they are dealing with the same problem as every macroeconomic model - the economic conditions created by the coronavirus violate basic assumptions of these models, causing them to give less accurate and even unexpected results. I think of it like driving a car with a broken speedometer: you have an idea of how fast you are going and if you speed up or slow down you'll have a rough idea by how much your speed has changed. The takeaway here is that unemployment was higher and came down because we added jobs. We can estimate the range for each of those figures, but don't know the exact number. That is still good news. If you are one of my readers who follows this more closely and think I've missed something, let me know.

The most current figures show Houston with about 452,000 unemployed and a 14.2% unemployment rate for April. The May figures won't be available until July 1.

SO WHAT DID PEOPLE DO WITH IT? PART 2. Last week I ran a story about how government transfer payments and reduced consumer spending was boosting US personal savings, and quoted some evidence that suggested people were investing it the stock market. I feel pretty solid about that still, but some new evidence came out this week from the Federal Reserve that showed that's not the only place it went.

The Fed's H.8 Assets and Liabilities of Commercial Banks in the US dataset showed that credit cards and other revolving loans have dropped by about $68 billion, an almost 10% change - and only the second decline since 2000. There's a negative view to this story if the decline reflects a long term retreat by consumers. Certainly a decline in revolving credit would have been expected after the decline in spending we noted last week. But there's another part of the story that's positive. The second chart shows the trend in MBA Purchase Index for new home mortgages After starting 2020 strongly, the index fell to 2014 lows in March. It has returned and is now higher than any year since 2011. Perhaps not all of that money went to Wall Street - perhaps at least some of that went to clean up account balances and credit problems in advance of buying a house.

BLOOMBERG PREDICTS BIG SECOND WAVE OF JOB LOSSES. Bloomberg Economics ran a story this week projecting as many as 6 million high-wage job losses coming in a second wave. It's entirely possible t he losses projected in this story may be accurate, but there was one critical line in the explanation of their methodology:

"The economists then took government data on relations between industries to compute the ones most reliant on demand from the most-affected sectors. Combining that information with the hit to employment in the most affected sectors they extrapolated to other jobs at risk, most of which were higher-skilled, white-collar roles."

That means they are using the  BEA's Regional Input-Output Modeling System . The  BEA RIMS II documentation  includes a lengthy warning and explanation that RIMS II may not accurately forecast under COVID-19 conditions:

"RIMS II multipliers are not designed to estimate the impact of supply shocks to the economy. RIMS II makes several assumptions about how businesses and households operate. These assumptions work well when fundamental relationships and structures remain stable in the economy. However, during periods of economic instability, such as during the COVID-19 outbreak, these assumptions may no longer be applicable."

This model is a reasonably good one and I've even featured it here in a past issue to perform a very similar calculation - in this case the financial impact of job losses on the Houston market. However, my story about the model included a warning about its limitations, potential for error, my low confidence in its results, and that it was used to indicate only a broad scope and direction. Bloomberg's projection involved significant economic impact and should have included similar warning.

All charts and data from Houston Realty Information Service.

IF YOU READ NOTHING ELSE, READ THIS. STAT has done a really good piece on what to do now so we don't screw up the virus response again.

SIGNIFICANT RISE IN CASES. I've included the SETRAC dashboards and the Texas Medical Center COVID-19 Early Warning Signs, based on the guidance provided by Governor Abbott. Cases have started growing dramatically in the last week and the TMC says current trajectory suggests base ICU capacity could be exceeded in two weeks. At current capacity, ICU hospitalizations would need to double to reach that level; the TMC alone also possess an additional 1,200 beds of surge ICU capacity. We will need to watch this very closely and should expect at least discussions about a pullback of some measures to see how cases respond. As I have mentioned before, research shows 99% of cases show symptoms within 14 days. Honestly I am not sure what the next steps are in virus mitigation at this point.

NOT THE FIRST TIME CURES HAVE BEEN POLITICIZED. You may have recently seen the story about the retraction of a major study criticizing the use of hydroxychloroquine. The retracted study is what led the WHO to end trials of the drug all over the world. I really have no opinion about the study or the drug except that it has become yet another politicized issue. That's tragic, but it's not the first time this has happened in America.

In 1793, Philadelphia was ravaged by an outbreak of yellow fever - it killed about 10% of the city's population - and the nation's nascent political parties coalesced around cures promoted by two signers of the Declaration of Independence. Dr. Benjamin Rush represented the medical establishment and preferred treating patients with mercury and leeches. He was supported by the Democratic-Republican party. His opponent was Alexander Hamilton, Secretary of the Treasury and a Federalist, who supported using the bark of the cinchona tree - ironically the natural source of quinine and ultimately hydroxychloroquine. Hamilton had survived the disease and credited the bark as the cure.

The parties and their affiliated newspapers spread misinformation about cures, only presenting the research and findings that were supported by their partisans, and were sharply divided over whether the infection arose from within the US or was imported from foreign shores. Ultimately neither cure worked and the outbreak ended when winter snows killed the mosquitos that spread the disease. A vaccine would not be made until 1938. But the political division over the disease in 1793 hampered efforts to develop an effective treatment. You can read more about the dispute here.

D-DAY ANNIVERSARY. Saturday is the 76th anniversary of the D-Day landings in France. In commemoration of those who sacrificed to achieve victory, I share this item:

Some viewpoints worth reading:

Are you looking at new deals? Want to understand what's going on in the market today? We're here to help, even if it's just a short conversation.

Stay well.