Last updated 3/30/2020 at 8:17 PM PST.
I find myself in a situation where I did not cut losses before the markets took a huge downturn due to COVID-19. While there are no transaction costs involved for me to do so, I decided not to. Now, another way to mitigate this loss is to invest at the very bottom (or close to it). However… how can we tell where the bottom will be given the volatility of the market?
Let’s start by looking at the 2-month price performance of SPY for starters. More specifically, from 1/25/20 to 3/25/20. Notice how the market goes from non-panic mode (1/13-2/18) and swings to full panic mode.
Here’s my take on the matter. The market indices will continue to decrease for the following reasons (in order of importance):
- Quarterly reports: Companies will release their Q1 reports sometime in April. We can expect most companies to be negatively affected by COVID-19 because:
– Disruption of work (having to work from home or change the way they work).
– Employees becoming sick and being unable to work (especially in service industries).
– Disrupted supply chains.
With the inevitably poor Q1 reports, we can expect poor performance in those companies and thus the market indices. We will probably see significant effects into the second quarter.
– Schools have been partially shutting down and moving to online classes. CSU Long Beach, for example, has stated that in-class sessions will no longer resume for the rest of the semester (after being suspended on March 12th).
– Events are being cancelled, postponed, or altered to prevent spread.
– People are panicking and acting differently. We can see this with stores running out of stock on many items, and even in the market itself. 10-Y Treasury rates fell drastically to .54% on March 9th, indicating that people were flocking to safer investments (higher demand means prices increase and rates fall).
– The Federal Reserve has been cutting interest rates to help businesses weather the effects of the virus. As of March 15th, the Federal Reserve cut interest rates down to near 0% for the first time since the 2008 financial crisis.
– A government stimulus package is being negotiated to help prevent or assist people during a recession, which is seeming more likely by the day.
– As far as technical patterns go, this looks like many dead cat bounces or bear market rallies.
- Exponential growth of COVID-19:
– This video summarizes the exponential growth model and concern surrounding COVID-19. One notable takeaway from the video is that once people start to panic (worry), that’s actually a good thing for the markets in the future, as people are panicking from other people not panicking. In other words, now that people are worried and taking more precautions to prevent the spread, this will reduce the spread.
– Once the growth factor hits close to 1 and start to decrease, then we know the epidemic (in the U.S.) is starting to end, given all else is equal.
I created this Excel chart with some data on the current growth factor and current infected totals in the United States. I will keep this as up-to-date as I can.
While I can’t say for sure when the markets will bottom out (of course), it’s likely they will bottom out sometime around mid-April:
-given that the growth factor will have hit around 1 and start to decrease as well. If the growth factor hasn’t hit that point (its inflection point), then the bottom isn’t likely to have happened yet.
-given that the VIX doesn’t continue to increase or remain high around mid-April.
-given that unemployment doesn’t increase drastically by mid-April (which is looking like it will, unfortunately).
-given that the government stimulus response is sufficient in helping those in need.
-Another thing to consider is that the last time the Federal Reserve cut rates to near 0% (0%-.25%) was December 17th, 2008. What happened roughly 6 months following that move was the lowest rate for SPY since 1996. If we assume that this is a similar scenario, we can expect markets to bottom out around September 15th given that historical scenario. I have a feeling this is more realistic in the long-run, as it appears the economy will take quite a hit from businesses being forced to shut down and people being laid off.
I’d say it’s fairly certain to expect even more price drops. However, markets don’t always react as we expect to available information, and often times, it’s quite neurotic (see VIX). There’s still a lot of uncertainty surrounding this virus, so markets will likely continue to fluctuate until quarterly reports, more virus data points come in, and government intervention. As we near mid-April, we can see if the market will truly reach the bottom or if a recession will cause it to fall more. Stay tuned!
The growth factor appears to be decreasing (which means the spreading of the virus in the United States is slowing down), but I can’t tell for sure until tomorrow’s numbers come in. I’d like 6 points of data at least before I can confirm a trend. The data appears to be stabilizing as more tests for the virus are conducted and more people are diagnosed. My concern is whether the markets will react to this information or not, and the degree to which the economic loss will impact the market later on in Q2.
I was wrong in the past update about the potential growth factor trend. The growth factor is not decreasing anymore. It actually went up given Tuesday’s infected numbers. This means that the cases are increasing a lot still.
There is also talk about a stimulus package, increasing unemployment numbers, and an airline bailout. I’ll look at these when more information comes out in the following days, but it doesn’t look good.
The growth factor trend is not looking good. You can see for yourself above.
Some other things to look at in these uncertain times are the VIX (volatility index) and some Google Trends data.
As you can see, the VIX is showing us that volatility was reaching 2008/2009 levels. When we combine this with current speculation of high unemployment in the coming months (and Google Trends), it paints a fuller story that is reminiscent of 2008.
The situation is about the same as it was yesterday. I didn’t notice anything big changing. I decided to look into forecasting, given the current numbers. I’ll update this as well as we get more data.
This chart is a scenario where I took the current average growth factor and held it constant. I figured that this would be a more conservative model as opposed to a 6th degree polynomial regression or assuming that the growth factor would continue to increase (which is not necessarily true). These are the projected infected numbers from my method.
Assuming the growth factor does not decrease, we can expect these kinds of numbers in the following weeks, if my model is fairly sound. This will, of course, change if the actions we take as a nation influence the amount of newly infected people on a daily basis.
Additionally, states are starting to prohibit people from going outside except for essential needs and activities. I expect this to take hold in more states as the days pass. Hopefully we get updates over the weekend from the federal government about the anticipated stimulus package.
Today was the first day I could test my infected number projections. Unfortunately, the conservative nature of my model has shown through. The reported infected numbers today were higher than my model predicted by 830 cases. The slope of the growth factor has also increased by 15.6%, which means that the rate at which people are getting infected is increasing. Although, this could be due to more people getting tested, I’m not sure. Another thing to note about the growth factor is that it’s been decreasing in both Italy and Sweden.
The stimulus plan is also making headway in congress. So far, it appears that it will address needs of the workers, small businesses, and unemployment insurance, according to Senator John Thune. We’ll see what it really covers when it’s finalized and released.
Unemployment benefits searches have been increasing.
I’ll monitor how far the actual numbers deviate from my projected numbers and adjust the model accordingly if I need to. I decided to go with a 4th degree polynomial model for the time being, as that yielded results closest to the reported numbers.
A disturbing thing to note is that at the current rate of growth of COVID-19 in the United States, it will surpass Italy’s infected cases in about 7 days, according to my 4th degree polynomial model of growth. See below. I expect the market to really tank if this happens and it goes mainstream. There are some flaws with comparing cases to Italy, as it’s a smaller country, but it’s still significant.
I’m currently working on making separate posts for separate topics, thus making COVID-19 and its implications easier to explore. I understand it’s a complicated mess! This might take a while given online classes are back in session for me, and it’s a lot of information to organize. Here’s a brief summary of my observations on what has happened in the past few days:
– Looking at SPY from its peak before 2008, to after the 2008 crash, a 56% drop occurred. At the time of this post, the current difference between the SPY peak and current bottom is 34%. While things have been calming down (VIX has been going down and things are making some progress in congress) I still expect the market to go down. S&P500 futures are still down as well, so this tells me that the bottom isn’t too soon. In fact, I could tell that Monday was probably going to be bad for SPY as well because of the futures trading on Sunday! The futures didn’t go down a lot on Sunday, but neither did SPY.
– Perhaps I’m a little late to the boat on this one, but current short-term price performers are Zoom (ZM), Slack (WORK), Citrix (CTXS), and Microsoft (MSFT). Although, I’d be cautious about long-term price performance because of the recession and other problems I notice about these companies. However, it’s very likely they will be bullish for the short-term. Any internet-based stock is likely going to do well in a similar manner.
– Since some kind of recession seems inevitable, the question some people ask is how fast we will get out of it. I read some articles about economists suggesting that it will be a “U-shaped recession,” which means that economic impacts of COVID-19 will likely last several quarters and take a long time to recover. I agree with this sentiment. The Philadelphia Fed business conditions index showed a 12.7% decrease in March, from a 36.7% increase in February. Jobs are also being lost in the service industry. Treasury Secretary Steven Mnuchin also stated that lockdowns will likely last 10-12 weeks. It’s inevitable that many businesses will suffer greatly from this. Again, I’d like to go into deeper detail about this economic phenomena in a separate post.
– Congress isn’t getting anywhere yet with the stimulus bill. They can’t agree on what to include in the bill. Republicans accuse the Democrats of putting partisan items in the bill, when they should keep it out and focus on the main issue. You can see the main sections of the bill here. This bill is so important because I bet the markets will react differently based on what gets passed in it, and if investors deem it “enough” to support the economy and businesses. Bad bill = bad press = uncertain/scared investors = more market downturns. I’d also like to go over this in a separate post, especially when it’s finalized.
– I also saw this Fortune article that claimed Goldman Sachs and Morgan Stanley both have rather high estimates of GDP drops (30%).
– If you’re in the U.S., this resource gives some urgency to the dire situation. It’s an estimate of hospitalizations vs. how many beds are available. Using this in conjunction with my infected estimates (and then a % of people who become hospitalized) would be useful in determining whether our medical system will get screwed or not. I might dive into this more later as well.
– S&P500 made gains two days in a row. This is likely a rally behind the stimulus bill which will probably finalized and passed Thursday 3/26. S&P500 futures are also up a little bit. However, I would caution against buying at the moment. It may seem like the worst has passed, but it’s very possible it hasn’t. The coronavirus is slowly dying down but the economic impacts are still yet to be seen (GDP, company financial reports, unemployment numbers, etc). President Trump also proposed getting businesses back up and running as soon as possible, but I fear this may cause another jump in cases if he doesn’t give enough time for the virus to burn out. I also found this nice graph showing a possible pattern that should be considered.
– Unemployment benefits searches increased again after going down for a while. This could be due to the stimulus bill getting close to passing (and people are trying to search for new benefits). It could also be more people being let go from work.
The unemployment stats came out today. This is what they look like.
Despite this data, the S&P500 and its futures are still up. It appears that investors are satisfied with the stimulus package so far. The question is, will this stimulus package be enough to prevent a recession?