Download PDF: Monthly_202004
With the COVID-19 already spreading globally for over two months now and stock market seems to have bottomed, investors are right to ask whether we are at the peak of COVID-19 infection rate. In this article, we discuss where we are in the COVID-19 crisis cycle, the implication on countries’ banking system and asset prices, and potentially attractive investment opportunities.
First, new community infection in Mainland China has declined to zero in the past days and new cases are arising only from travelers coming back home. New confirmed cases in Europe (Italy, Spain, Germany and France) seem to also peak out already, although the death number will likely keep increasing in the coming weeks. Hence, it is right for investors to scrutiny the number coming from U.S., where state lockdown is still on its early period and the infection number has yet to peak daily.
The good news is that workers in China has started to work and the government has relaxed some quarantine rules to support the economy rolling again. Coal consumption has been back to above 70% of its normal and numbers of traffic congestion and pollution have been rising gradually, a good sign for the manufacturing sector. However, all this good news is still weighed by demand slowdown from outside China, which is a tailwind for the export-oriented economy. Going forward, we expect things to get better in Asia.
What is the data saying? Now, let’s talk about numbers. Regarding the disease mortality rate, we could look at Iceland case. Iceland tested 27.000/1M people, much higher than Korea (5.000/1M) and is the highest in the world. If we take Iceland sample as the best proxy for the disease anywhere, the result shows that infection rate among population is 2.16%, with a case fatality rate of 0.28%. To support Iceland’s data, there is also one town in Italy that got all of its 33.00 population tested and found 3% infection rate. This finding is significant in two ways. First, much more people than reported might already have been infected and showing no symptoms at all, which is bad because asymptomatic people could be spreading it to others unintentionally. Second, if the statistic is true then the disease is far less fatal than estimated earlier (2-3% rate in Hubei and 1.4% on cruise ship case), although being 10 times more deadly than common flu.
How does this compare to previous pandemic? Using the data obtained from CDC website, we compare the COVID-19 with swine flu pandemic in 2009, which has infection rate in U.S around 18.6% (60.8m/327m) and case fatality rate around 0.02% (12.469 death/60.8M case). This translate to swine flu being 10x more infectious, but also 1/10 less fatal compared to the current COVID-19 case. It is likely that when all this over, the death toll globally will be similar to swine flu pandemic, but the age bracket of death being different. More younger people die during the swine flu, whereas for COVID it is older people who died, mainly because 1/3 of old people has antibody against the swine flu from previous flu season.
How will the crisis play out from here, especially in the U.S? In our mind, there is two possible scenarios, with the former more likely and already is taken by many countries and several states in the U.S.
- A national lockdown, the faster the better, would be the most prudent way to slow the virus spreading further, at the expense of huge economic cost. First, lockdowns in China and several European countries have been proved to work if implemented early. With monetary and fiscal support for companies to stay afloat for another month (a big if), economic activity will resume normally once the quarantine is relaxed.
- If we applied the number discussed above, a 2-3% infection rate and 0.2-0.3% case fatality rate, it’s possible that the death number in U.S. will reach 12.800-28.800 people in the lockdown case.
- Current approach: have only few state lockdowns, do nothing for the rest and let the virus run its course. The lockdown in only several states is arguably useless, as cross-border movement will eventually cause the virus to spread nationally. This not to mention that the health care system will be overwhelmed, causing much higher death rate, possibly comparable to those in Hubei/Italy.
- Based on 5-20% of U.S. population got flu every year and a range of 1-5% fatality rate, death toll in U.S. could be between 160.000 and 3.200.000 this year. This is not accounting that the virus could potentially still be alive next winter, infecting more people that hasn’t been infected this year.
- Using such wide estimate of death, this means that if the quarantine measure costs $5 Tn (25%) of U.S. economy and no quarantine means no economic damage and let’s say 160.000 people died, then for every person that didn’t died because of the quarantine measure, the cost is roughly $ 31 million/person. On the higher death estimate (3.2 million death), the quarantine measure translates to $ 1.5 million economic cost for every life saved.
Will Emerging Market Countries Face a Banking Crisis?
In the past decade, many risky companies and countries have been on a borrowing binge on the backdrop of rising stock prices, low volatility environment and low bond yields. With trillions of bonds trading at negative yield, debt financing cost has been suppressed globally and demand for higher yielding dollar debt has been on the rise. This supported many countries and companies in Emerging Markets to borrow in hard currency, which tend to be cheaper, exposing themselves to the risk of currency mismatch between revenue and cost, and increasing their debt burden significantly when their domestic currency depreciates relative to the hard currency (U.S. dollar).
On the table below, we identified EM countries (and companies) with highest growth rate of dollar borrowing in the past decade. The result is not very encouraging. Despite the slower post-GFC growth across EM and few countries manage to double their economy, many countries (and companies) such as Indonesia, South Africa, Chile, India and Mexico have more than doubled their borrowing in hard currency. In the current rout, these countries’ currency, except Indian Rupee, have collapsed between 10-25% range, increasing their interest burden at the same time when their revenues are likely collapsing.
Table 2 shows that last year, South Africa, India, Chile and Mexico have ramped up their dollar debt borrowing, making them a potential hotspot for a banking crisis from the domino effect of recession due to much weakened global demand. NPL will inevitably rise in the coming months as companies face tightened liquidity in accessing debt/loan and cash flow will be much lower. The good news is that capital ratio of the banking system in these countries are adequate and governance has been tilting to a more conservative stance in Indonesia, Mexico and South Africa in the past two decades.
What About Dollar Liquidity Impact to Developed Countries?
EM countries are not the only participant in the last round of dollar borrowing. Developed countries (ex-US) dollar borrowing far exceed EM (3x), in terms of amount, share of credit and GDP. Hence, a squeeze in dollar liquidity will hurt DM companies’ funding greater than those of EM. But at the same time, these DM countries have swap lines with the Fed, whereas EM central banks don’t.
In a big picture, however, the dollar borrowing share as % of total credit to non-bank sector is relatively low at 16% for DM Ex-US and 6.6% for EM. Within EM, the more worrying part is slower growing countries who have been borrowing aggressively in dollar of late, as discussed above.
Table 3. EM and DM (Ex-U.S.) Dollar Debt Borrowing
Where Is the Investment Opportunity?
The answer is everywhere! Some key points:
- Although China has been outperforming and its economy is leading the recovery, it is possible that other EM countries will catch up in the next 3 months. The quarantine will likely be over globally in mid-late April and battered-down countries equity will jump higher when infection rate starts to decline.
- The market I have falsely like in the past, such as Turkey, Mexico and Chile have done among the worst in this rout. These are the riskiest market in EM, based on their dollar debt in among corporates. However, all of them now have single digit P/E ratio, with Turkey and Chile trading at 0.6x and 0.8x book value, and Mexico at only 1.1x (which is a diversified market with favorable demographic growth). The obvious risk is that these countries will have a banking crisis, the reason that authorities in these countries have been slower in acknowledging the virus spread and implementing lockdown. In this light, the difference in drawdown among EM countries somewhat makes sense. The cheapest one is the one with highest dollar debt, hence more likely to have liquidity/solvency issue arising from sudden stop in economic activity.
- Currently, further downside risk on EM assets lies not in the stock prices itself, but rather in the currency. Central banks across EM, even those with already low rates (Brazil), are cutting rates when dollar liquidity is evaporating. This has resulted in currency collapsing and more difficulty for corporates with dollar debt, creating a feedback loop between corporate and banking system (NPL go up and credit will be harder to obtain, causing further NPL spike). Unlike in U.S., EM central banks are not buying corporate debts in primary/secondary market, hence there is no backstop in EM countries for corporate to obtain liquidity.
- Table 4 outlined the % correction of EM equities from its recent high, their valuation metrics and where they are compared to history. We summarize our view at the end of this section.
Chart 3. EM countries equity attractiveness from global perspective
We know that valuation has improved significantly across the board, but such metric is a poor market timing indicator. Few people could time the market correctly, but if history is any lesson, then the current market movement is consistent with the low during 2008 GFC, with treasury return peaking out slightly after equity bottomed, a combination we are observing currently.
- We like Chile for speculating on upside potential of copper prices. Equity multiples are cheap and Chilean stocks, historically strongly correlated to copper prices, have lagged even before the market rout.
- Maintain long position on Turkish equity. P/B value is only 0.6x and the country has undergone a massive shock in 2018, cleaning all the excesses. Recovery will likely be faster than the rest of EM and the currency has cheapened substantially.
- Mexico offers a diversified market at a historically low valuation, in both P/E and P/B terms.
- Speculative long position on Russian equity, in anticipation of positive shock from Russia-Saudi deal on limiting oil production. Valuation of Russian equity (0.8x book value) has reached its historical low and downside risk is limited.
Copyright © 2020, Putamen Capital. All rights reserved.
The information, recommendations, analysis and research materials presented in this document are provided for information purposes only and should not be considered or used as an offer or solicitation to sell or buy financial securities or other financial instruments or products, nor to constitute any advice or recommendation with respect to such securities, financial instruments or products. The text, images and other materials contained or displayed on any Putamen Capital products, services, reports, emails or website are proprietary to Putamen Capital and should not be circulated without the expressed authorization of Putamen Capital. Any use of graphs, text or other material from this report by the recipient must acknowledge Putamen Capital as the source and requires advance authorization. Putamen Capital relies on a variety of data providers for economic and financial market information. The data used in this publication may have been obtained from a variety of sources including Bloomberg, Macrobond, CEIC, Choice, MSCI, BofA Merrill Lynch and JP Morgan. The data used, or referred to, in this report are judged to be reliable, but Putamen Capital cannot be held responsible for the accuracy of data used herein.