Houston Coronavirus Tracker Update 7/25/20

SCHOOL CLOSINGS MEAN FEWER CLOSINGS. It seems like every time I go to launch the newsletter something new happens. With today's (Friday) announcement by Judge Hidalgo that schools will start virtually until September 8, I think we can expect sales over the next week to drop. In week 13, the March 24 SAH order, we saw sales drop by 50% before recovering in week 14. Week 26 included Gov. Abbott's order to reclose bars and other restrictions we saw about a 30% decline. I'm not sure that we will see as large a drop after this announcement since it is theoretically time-bound, although many buyers will suspect this is likely the first interval of a rolling closure. Nevertheless, I would expect sales to be off on year-over-year comparisons as they have been with similar news and to mostly, if not totally, recover the following week.

HOSPITALIZATIONS. As I've discussed over the last few weeks, we are now starting to see what looks like a peak around July 15 in both ICU and general isolation hospitalizations. There's an even more pronounced pattern in cases, but I didn't include this because I still think cases are an unreliable metric. I think this is also reflected in the shift in media coverage to the problems in the Rio Grande Valley. I think the next big challenge will be increased fatalities - which should follow the increase in hospitalizations by 3-5 weeks. Just know that this is a lagging and not a leading indicator. The most reliable indicator is what you see on the chart below, and if you'd like to monitor them yourself, you can follow the SETRAC dashboard.
  If you liked testing, you’ll love the sequel – the vaccine.  Two vaccines, from Moderna and Oxford-Astra Zeneca have shown to produce an immune response in humans. This is great news.  It looks like the US has ordered several different vaccines totaling around a billion doses.  It’s uncertain exactly how many people must be vaccinated to achieve herd immunity; for measles, it’s 95%, but it’s likely to range between 80-95% . I was originally concerned about having enough doses:  t here are 330 million people in the US . But here’s the challenge: we don’t know which vaccines will emerge from trials, and it’s possible that we’ll need more than one type of vaccine. But the leading candidates -- both the Moderna and the Oxford-Astra Zeneca vaccines require two shots – meaning the 300 million doses we’ve purchased of Oxford-AZ, for example, will only cover 150 million people, or just under 50% of the country.  
How would vaccinations work? To inoculate everyone in America in 30 days would require 1.1 million shots every hour (10 hours/day). If you shot a person every three minutes and had 5 people per location, that’s about 11,000 locations – or slightly higher than the total number of either CVS or Walgreens stores in the US. A big number, but doable if you have the doses and the staff – remembering also this will need to be done twice at some currently unknown interval. Takeway: although there should be enough doses to inoculate everyone eventually, expect to see as much chaos in vaccine administration as we’ve had in testing – particularly when limited doses may require some sort of priority or triage determining who gets a shot.
Lastly, I don’t think that people understand it’s likely to take several weeks after the shot/s for immunity to take effect – so I expect governments will at least attempt to prolong infection control measures until a majority of people have received (a) both shots and (b) a sufficient interval to acquire immunity. Takeaway: this won’t be over until a long time after the vaccine is approved and starts being administered.
There’s one more thing about the vaccine:  according to Bloomberg at a UK parliament hearing this week Jonathan Van-Tam, a deputy health minister, told the committee that he is more optimistic about a vaccine being ready by Christmas, but in any event it would not be possible to administer it close to the flu vaccine . “We won’t have any data on whether it can be given at the same time,” Van-Tam said. “Likely we will have to delay or separate those two vaccines apart in terms of the staging we give them to the relevant patients.” Takeaway: it’s unclear what they mean by “delay,” but what is clear is that we should expect more of it.
I'm sorry, this is "Doing it right?"  'Extreme dysfunction': New York City dropped $52M on a coronavirus hospital that served fewer than 80 patients.  Says a nurse who worked there: "I basically got paid $2,000 a day to sit on my phone and look at Facebook. We all felt guilty. I felt really ashamed, to be honest.” Now I have been fairly critical of the local and state response to the coronavirus here, and if you talked to me in person it would even be worse. But I will say this for Texas; if we were our own country our deaths per capita (148/MM) would rank #32 – between Krygyzstan and Bermuda.  To what countries does New York compare? None. Both New York and New Jersey have per capita fatality rates 30% above the every country on the planet. While New York let the facility in this article, the Javits Center and the hospital ship Comfort sit virtually unused, the state was saving hospital capacity by sending infected patients back into nursing homes, where 6,497 residents died of COVID-19 (and another 6,732 in NJ). That’s more than everyone who has died from it in Texas, as of this writing.

The chart below shows death rates/state (per 100K instead of per 1M). By my count Texas is #30. We would probably even better if it were not for the absolute trainwreck the state has made of contact tracing.
On the list of ways to detect Coronavirus, this is my #2.  In my exhaustive search to flush out the best way to find out how widespread the coronavirus infection is, testing at one method keeps floating to the top. Everywhere WWTP testing for viral shed is tried it turns up evidence of infection spread orders of magnitude greater than regular testing, and has been shown to predict caseloads 7 days before with an R 2 around 98%. If I could predict your sales with that accuracy, you’d never hire another consultant again.   Now the State of Colorado has begun testing sewage effluent for viral spread to gauge the extent and status of infections.  This would let you pinpoint measures even to small areas, and give you almost immediate feedback on their effectiveness. Our current methods of measuring the spread of infection are like estimating the number of people who are speeding based on how many tickets are written. To sum up best ways to find out how many people are infected:
  • #1 – random sample antibody testing
  • #2 – wastewater treatment plant testing
  • #3 – any other conceivable method
  • #4 – counting the people who got swabbed at Delmar stadium because their best friend's sister's boyfriend's brother's girlfriend heard from this guy who knows this kid who's going with the girl who saw someone with the coronavirus at 31 Flavors last night.
SCARY NEWS FIRST. Almost all the economic news right now is positive or improving, except unemployment which is persistently high. While I don't think that we have very many buyers relying on PUA and it shouldn't necessarily represent a threat to home sales -- it could put significant pressure on the rest of the economy if this replacement income goes away before companies are ready to hire again. And generally, the economy really dislikes uncertainty of any kind.
#1 SAVINGS GOAL FOR YOUNG AMERICANS IS A HOME. A recent survey from Travis Credit Union seeking to learn more about the money-saving habits of young Americans (born 1981 - 2012, average age 28) and how Covid-19 and the looming recession has impacted their savings, polled nearly 2,000 Millennials and Gen-Zers and what they found is shown below. The survey also found that about 40% have had to go into savings to survive during the pandemic, consuming on average 33% of those savings. Still, 3 out of 4 remain optimistic about their financial future.
ESCAPE FROM NEW YORK AND OTHER CITIES. The chart below indicates urban landlords are having a much more difficult time collecting rent, while suburban properties do not have nearly as hard a time collecting. Just another datapoint about the viability of suburban development in the post-pandemic world.
RESTAURANT RECOVERY FINALLY? The chart below shows the change in restaurant visits versus a baseline - in early July in Texas and other states with severe outbreaks you can see they have started to recover. There has been some correlation between restaurant traffic and case increases, but if this pattern is true, it appears that perhaps that relationship has broken down, at least for Texas.
DEFAULT PROBABILITY HAS GONE WAY DOWN. The chart below shows the probably of default by companies in the most severely impacted industries.

  • Home sales in July show another month with an increase over last year despite the high rises in cases that we saw in the first two weeks.

  • Weekly sales show that same pattern of recovery, with really just a handful of weeks responsible for declines - and specifically weeks tied to policy changes related to the pandemic.
Average closing price sustained a raising pattern and that shows home market is doing well despite the crisis.
Pending home sales remain unchanged.
Last month terminations had a huge decline compared to last year - as we saw a huge rise in virus cases again suggesting it's the policy response and not the pandemic that's responsible for sales declines earlier in the year. Terminations have increased slightly in July, so we'll have to see how they shape up through the end of the month.
On the other hand, home rentals are lagging slightly behind last year for the first time since April. July is usually the highest month for home rentals in the MLS as families look to get settled before school starts. It's possible the uncertainty around in-person vs. online school may have relieved pressure in this market this summer.
Some viewpoints worth reading:
  • Work from Home trends decreased traffic by 270 million miles/yr
  • Increasing COVID-19 caseloads were associated with countries with higher obesity, median population age and longer time to border closures from the first reported case. Increased mortality per million was significantly associated with higher obesity prevalence and per capita gross domestic product (GDP). Reduced income dispersion reduced mortality and the number of critical cases . Rapid border closures, full lockdowns, and wide-spread testing were not associated with COVID-19 mortality per million people. However, full lockdowns and reduced country vulnerability to biological threats (i.e. high scores on the global health security scale for risk environment) were significantly associated with increased patient recovery rates.
  • Models missed … again. On May 24th, Imperial College-London issued a forecast for coronavirus fatalities as a result of reopenings in five US states. States that peaked back in March and April show no signs of a resurgence, let alone the predicted resurgence that would surpass the first wave. And states that are undergoing later surges are still well below the ICL team’s predictions – so far below that they barely even register on the graphs. These flawed ICL models begin with an unproven assumption, namely that lockdowns are effective at combating the coronavirus. It's important here because it could be claimed before the outbreak that model recommendations prevented fatalities. With the openings, however, we know that mobility met or exceeded the model assumptions while the fatalities we've seen are a tiny fraction of what the models projected. That tells us that the the models are therefore automatically calibrated to produce a sharp spike in deaths after the removal of lockdowns or any move toward reopening. Their results should therefore be interpreted accordingly.

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.

Scott Davis
Location Strategy, LLC