Saturday, February 11, 2023

Data Update 4 for 2023: Country Risk - Measures and Implications

I describe myself as a dabbler, and it does get in the way of my best laid plans. A few weeks ago, I posted my first data update pulling together what I had learned from looking at the data in 2023, and promised many more on the topic. In the month since, I have added two more data updates, one on US equities and one on interest rates, but my attention was drawn away by other interesting stories. Thus, I took a detour to value Tesla, around the time of their most recent earnings report on January 26, and added a second post to respond to the pushback that I got. About a week and a half ago, just as I was getting ready to start on my fourth data update, I got distracted again, this time by a story of a short seller (Hindenburg) targeting one of India's most visible companies (Adani Group) and I don't regret it, because that story is a good lead in to talking about country risk, which is the topic of my fourth data update. Irrespective of whether you think Hindenburg's short selling thesis against the Adani Group has legs, it is undeniable that the fate and value of this family group's companies is intertwined with the India story. A strongly growing India needs massive investments in infrastructure to succeed, and the Adani Group seemed uniquely qualified because of its perceived capacity to deliver on its promises, as well as its political connections. 

Country Risk - The Ingredients

    At the outset of this discussion, it is worth emphasizing that there is risk in investing in every country in the world, with the differences being one of degree. Thus, you would be making a mistake, if you assume that this discussion only applies if you are investing in India, Brazil or Belarus, and that it does not, if your investments are in the United States, Germany or Australia. The developed/emerging market divide was created by practitioners as a convenience, and while it sometimes has consequential effects, as is the case when a company is reclassified as developed from emerging, or vice versa,  much of what I will say about how governments, legal systems and regulatory frameworks can affect corporate value applies to all countries.

Determinants

    If you accept my premise that not only is it more risky to operate in some parts of the world than others, but also that risk varies across countries and time, the next question become one of deciding what determines the magnitude of country risk in a country. In my annual updates on country risk, I go through these determinants in detail, but the picture below summarizes the drivers of country risk:


It should come as no surprise that the determinants of country cut across all dimensions, with politics, exposure to violence, legal systems and corruption all determining country risk exposure. It is not required, but it is generally true, that countries that score poorly on one dimension tend to also score poorly on others, with countries that are most exposed to war and violence also having dysfunctional or nonfunctioning governments and courts.    

A Life Cycle Perspective

    When asked to explain differences in country risk around the world, it is unfortunately true that much of that categorization is lazy and overly broad, often centered  around geography, culture and race. Thus, Asian countries were viewed as incapable of reaching first-world status, until Singapore showed that this was not true, at least on the city-state level, and Japan established its falsehood, with explosive growth and prosperity in the 1970s and 1980s. The stigma of being a Latin American or African economy persists, but there are success stories in both continents. At the same time, there are others who argue that groups with shared cultural or racial identities are incapable of elevate their countries to developed status. That is nonsense, since individuals within these groups often become success stories in a different setting or economy, unencumbered by the systemic inadequacies of their own countries. I believe that any country is capable of being a "first world" country, if it works systematically at creating a system that is perceived to be fair, timely in delivering legal redress and blessed with a government that has the interests of its populace as its first priority. By the same token, a country that is viewed as "first world" can lose that status, if people start perceiving the system as unfair, legal systems filled with delay and waster and a government that becomes capricious in its actions, or worse.

    On the specific question of how much governments matter in determining country risk exposure, I am going to adapt a structure that I use to look at companies, the life cycle, and apply it to countries:
In my simplistic picture, you can see how as countries evolve through the life cycle, the influence of government and country narratives changes on investment success/failure:
  • Role of Governments: In younger economies, the influence of government is central, in both good and bad ways, since these economies are almost entirely dependent on growth, and a combination of good (bad) tax, licensing and regulatory policies by the government can an act as a growth accelerator (destroyer). As economies mature, the effect that governments have on companies will recede, at lease on overall growth, though tax policy can  still redistribute that wealth and influence business behavior. When countries decline, government attempts to stem or slow decline can make them relevant again, in good and bad ways.
  • Country versus Investment/Company Narrative: That structure explains why when investing in a company in some countries, you have to not only do due diligence on these countries, but also form a narrative for how these countries will evolve over time. After all, your investment in Dangote Cement, a cement company with a dominant position in Nigeria nd West Africa, will do much better if that part of the world does well and will be handicapped, perhaps even fatally, if there is political and economic upheaval. In contrast, your investment in Krupps is less likely to be affected much by your views on the German economy.
  • Uncertainty: When investing, uncertainty is part of the process, but when that investment is in a project or company in a young country, a significant portion of the uncertainty is about the country, rather than about the company or investment. Put simply, you are unlikely to find safe projects in risky countries, since country risk will undercut whatever perceived stability there is in the project's cash flows.
If your views on investing and valuation were formed by reading Ben Graham, and nurtured by listening to Warren Buffett, it is worth remembering the time and the setting for their sage advice. Put simply, the rostrum that when investing in a company, you should focus on the company's management and moats, and pay little or no heed to governments or macroeconomic indicators, may have worked for value investors in the United States, in the 1980s, but will not hold up not just in other parts of the world, but even in the United States in the 2020s. Globalization and the emergence of a world economy that is no longer centered on the United States has made it an imperative for all investors to think about and understand country risk. 

Country Risk: The Measures

    If investors have no choice but to deal with country risk frontally, in most parts of the world,  it follows that we have to come up with measures of country risk that can be incorporated into investment decisions. In this section, I will begin measures of country default risk, including sovereign ratings and CDS spreads, before moving to more expansive measures of country risk before concluding with measures of equity risk premiums for countries, a pre-requisite for estimating the values of companies with operations in those countries. 

Default Risk

   As with individuals and businesses, governments (sovereigns) borrow money and sometimes struggle to pay them back, leading to to the specter of sovereign default. Through time, these defaults have led to  consequences that range from mildly negative to catastrophic, with some defaults triggering invasions and political revolutions. It is also the aspect of country risk, where there is the longest history of measurement, and there are widely used measurement tools.

1. History of Sovereign Default

    In 2022, there were five sovereign defaults, with three (Russia, Belarus and Ukraine) a direct consequence of Russia's invasion of Ukraine, and Sri Lanka and Ghana joining the ranks, for different reasons. Those sovereign defaults are the latest in a long list of defaults that stretches back into the nineteenth century, and the graph below shows defaults in the most recent few decades, across geographies:

Source paper

For much of the documented history, Latin America has been the epicenter for sovereign defaults, though there has been an upswing in Africa in recent years. Looking at the defaults over time, it is also worth noting that local currency defaults (where a sovereign defaults on a bond denominated in the local currency) have comprised a sizable portion of defaults over time, as can be seen in the graph below:

Source paper

What does this all mean? The conventional practice, when estimating risk free rates, has been to use the government bond rate in the local currency, if available, as the riskfree rate in that currency, and that practice is wrong when markets perceive default risk in the sovereign and build that into the government bond rate. It is for this reason that I net out default spreads, based upon local currency ratings, from government bond rates to estimate riskfree rates  in multiple currencies at the start of 2023:

Download data

The differences in riskfree rates across currencies can be attributed to differences in expected inflation, which is at the heart of why a valuation that is consistent in its treatment of that inflation will be currency invariant. (Valuing a company in Turkish Lira should give you the same value as valuing the same company in Euros, differences in riskfree rates notwithstanding.)

2. Sovereign Ratings

    The ratings agencies that rate corporate default with ratings have also had a long history of assessing sovereign default risk, with sovereign ratings, with the numbers of rated countries increasing dramatically over time, with the number of countries rated by Moody's (S&) increasing from 33 (35) in 1990 to 152 (131) at the start of 2023.  Sovereign ratings, like corporate ratings, range from Aaa (AAA) in Moody's (S&P's) scale, to D (in default, with a couple of differences in how you read the ratings:

  • While each company generally gets one rating, countries are usually assigned two ratings, one for local currency borrowings and one for foreign currency borrowings.
  • The fundamentals that feed corporate ratings come primarily from its financial disclosures, though qualitative factors play a role. Sovereign ratings start with quantitative measures of a country's economic standing, but there are far more non-financial forces that seem to come into play.
No matter what you think about sovereign ratings as a measure of default risk, they are the most freely accessible measures of country default risk. At the start of 2023, I summarize the sovereign ratings for countries in the heat map below:

The red and orange part of the worlds have the highest default risk, at least according to Moody's, and you can see it covers large swaths of Latin America, Africa and Eurasia.
    While sovereign rating agencies have been accused of bias, with a skew towards giving lower ratings to emerging market countries, while over rating developed market countries, I believe that their real sin is that they are late in reacting to changes in default risk.  The last year (2022) was one that saw more bad news than good news on the ratings front, with Fitch downgrading 21 countries and S&P downgrading 16 countries (while posting a negative outlook, a pre-cursor to a ratings downgrade for 8 countries).

3. Sovereign CDS spreads

    The sovereign CDS market, a relatively recent entrant into the sovereign default risk game, has for the last two decades offered investors a market where they can buy insurance against default risk by sovereigns, and by doing so, provided a constantly updated, albeit noisy, measure of the default spreads of countries. In January 2023, there were 76 countries with sovereign CDS spreads available on the market, and they are listed below:


During 2022,  there was a suspension on trading on sovereign Russian and Ukrainian CDS, leaving us at the tender mercies of just the ratings agencies, It is worth noting that despite the abuse that ratings agencies get for ineptitude and bias, there is a signifiant overlap between their assessments and the market's assessments of country risk.

Overall Risk

    While default risk  measures are widely available and used, they can be rightly challenged as taking too narrow a view of risk. After all, there are countries that score low on the default risk dimension but are exposed to political and economic risks that are considerable, as is the case with much of the oil-rich countries of the Middle East. There are no easy remedies for this problem, but there are services that generate country risk scores that bring in multiple measures of risk. While the Economist, the World Bank and private services provide country risk scores, I will stay with Political Risk Services, a data service I have used for a long time, more because of my familiarity with it than for any perceived superiority in how it measures risk. The PRS reports risk scores for different dimensions of country risk, and a composite risk score, that includes all of them. The heat map below reports on PRS scores, by country, at the start of 2023:

Source: The PRS Group

More than in prior years, this year's PRS map reveals a divide between the default risk perspective on risk and the PRS perspective. For instance, India is viewed as marginally less risky than China, and both are viewed as riskier than Kazakhstan., and the United States is perceived as much riskier than Germany or the Scandinavian countries.

Equity Risk

    Having traveled the long and winding road from talking about the drivers of country risk to measuring country risk, we can take a shot at estimating the risk premiums we would use when investing in businesses, as equity investors, in these countries. Rather than bore with you the details of my approach to estimating equity risk premiums, which are described in excruciating detail in my paper on equity risk premiums (linked below), I will summarize how I estimated the equity risk premiums for countries at the start of 2023:


I start with the implied equity risk premium for the S&P 500 of 5.94% (see my second data update for 2023 for details) as my premium for mature market, and build up to the premiums for other markets from that, using default spreads as my starting point, and scaling them for the additional risk of equities. The resulting equity risk premiums, by country, are shown in the picture below:


With all of the caveats about country ratings and default spreads, the map still provides consistent estimates of equity risk premiums around the world. In fact, there are about a dozen countries that are unrated, where I have used their PRS scores to make estimates of their equity risk premiums.

Company Risk Exposure to Country Risk

   As a final piece of this post, I want to contest what seems to be the default assumption in much of valuation, which is that the risk of a company comes from where it is incorporated and traded, rather than where it does business. Effectively, it is what leads analysts to value US companies using the US equity risk premium and Indian companies with the Indian equity risk premium, even though both groups of companies may make their products in and derive their revenues from other parts of the world. I believe that a company's exposure to country risk should be based upon where it operates, though we can debate how best to measure this country exposure, with revenues, production or a mix of the two in play, for weighting. 

I know that there are risks that derive from where a company is incorporated, and that its regulatory and tax structure may be affected by that choice, but to argue that this is the dominant risk at play does not stand up to common sense.
    The implications for investment and valuation are simple. Investors and analysts who paint country risk with a broad brush, using country of incorporation to measure equity risk premiums, will over value developed market companies like Coca Cola, Apple and Netflix, with significant operating exposure in emerging markets and under value companies like Infosys (India), Embraer (Brazil) and Vinamilk (Vietnam) by assigning the domestic equity risk premium to them, even though they generate large portions of revenues from foreign, and often much safer, markets. 

The End Game for Governments

    I know that I am going against the current political trend, but I believe that the end game for a good government is analogous to that of a good founder, and that is, once it has provided the structure and the basis for economic growth and prosperity, it should make itself less central to the economy, not more so. Note that while this may seem like the libertarian position, there are significant differences.  I do believe that it is a government's role to craft laws and regulations that minimize the externalities that businesses create, but those laws/regulations should be few in number and changes, when they happen, should be reasoned and infrequent and enforcement should be fair and timely. There is nothing more unsettling than being a business person, consumer or citizen in a setting, where you are faced an avalanche of rules, sometimes contradictory, that are constantly changing, and enforced inconsistently.

YouTube Video

Data

Papers/Posts

No comments: