Lament to the government bond – 3rd quarter, 2018
On February 1, 2018, for the first time in its history, the Bloomberg Barclays US Treasury Total Return Index (formerly the Lehman Brothers US Treasury Total Return Index – moment of silence, please) delivered a negative return on a rolling three-year basisi. That is to say, if you bought the index exactly three years earlier, before adjusting for fees, taxes, and the decline in purchasing power of the currency in which your investment was denominated, your account balance would have been negative for the first time since some bankers at Lehman conceived the idea of tracking a basket of U.S. government bonds.
Bloomberg Barclays US Treasury Total Return Index annualized three-year rolling return
Jan. 31, 1976 to Sep. 30, 2018
Source: Bloomberg LP. Annualized three-year rolling returns in US$. As at September 30, 2018. The index measures US$-denominated, fixed-rate nominal debt issued by the U.S. Treasury.
The Bloomberg Barclays US Treasury index was designed to replicate a very broad basket of fixed-coupon U.S. Treasury bonds with the largest position constituting a 0.90% weight in the U.S. 2.875% note due May 15, 2028ii. Diversifying itself not by business idea but rather by term, the index has a small allocation to every U.S. government bond ranging in maturity from the 0.875% note due September 15, 2019, to the U.S. 3.00% note due August 15, 2048. No doubt it was this exposure to long-term bonds – whose price is the most sensitive to changes in interest rates – that led to the decline in value.
Negative nominal returns over an investment period of several years is a new phenomenon for government bond investors, and not one they should welcome. Persistently low interest rates and the low bond-interest payments that come with them mean returns from fixed-income investments are all the more likely to show deficit-red in the future. While a $100 bond paying $5 per year can lose $15 in value over a three-year holding period and still break even, the same security offering just $2.50 in annual interest has only half that buffer. And that’s exactly what’s occurred with the U.S. Treasury index shown above. Loaded with 10- and 30-year bonds paying 2.0% or 2.5% per year, coupons will do little to offset price declines in the underlying instruments.
Yet another dismaying data point for government bond investors is shown below, where the U.S. 10-year Treasury bond has delivered a negative trailing five-year return after inflation for the first time since at least 1980iii. More specifically, while the total return would have been positive, after having passed up a US$1.00 cheeseburger in 2012, a hypothetical investor might have invested in the “safety” of a U.S. 10-year Treasury bond, earned their 1.7% per year and walked away with US$1.08 at the end of a five-year investment period only to find that the aforementioned cheeseburger now cost US$1.09iv. What’s even more upsetting about this lousy outcome is that our investor had better get used to it. While the decline in nominal value noted in the Bloomberg Barclays index isn’t a permanent loss, investors in government bonds are in for a sustained period of negative real returns as yields that have remained so low for so long struggle to overcome the current rate of inflation. But we left out the best part – our hypothetical investor gets to pay tax on the US$0.08 in earned income, and pay a couple cents in fees as well. Gone are the days of risk-free income.
U.S. 10-year Treasury 5-year rolling return after inflation
Jan. 31, 1985 to Aug. 31, 2018
Source: Bloomberg LP. FTSE 10-year Treasury Benchmark On-The-Run Index and US CPI Urban Consumers Less Food & Energy were used to calculate inflation-adjusted returns for the U.S. 10-year treasury bond. As at August 31, 2018. September 30, 2018 data for US CPI Urban Consumers Less Food & Energy is currently unavailable.
The magic bullet…?
It’s unlikely that anything we’ve described so far is shocking to anyone reading this commentary; we all know how unattractive developed market government bonds have become in terms of absolute yield. We also know there’s no magic bullet to overcoming the hurdle of low interest rates in fixed income investing. But that doesn’t mean investors are not going to try!
In the investment world, things get popular. When things get popular, we get asked about them. And one of the things that got really popular really fast as developed market government bond yields became so abysmal was looking abroad to emerging market debt. Prospective EdgePoint investors especially will often ask about our capacity to look globally for higher returns in emerging market government bonds, often followed by some anecdote or story about the success of investing in emerging market fixed-income over the past three or four years compared to the experience in Canada or the U.S. Our answer is always the same: we don’t think we can add much value investing in government debt period, let alone emerging market government debt.
Our approach involves the careful study of a business’s prospects and growth potential, the quality of its management team, the threats to its business model, its moat or competitive advantage, and its future profitability, then ensuring we’re compensated appropriately given the risks involved in making an investment. Let’s call this process our assessment of a business’s “capacity to pay.” If we’re correct in our analysis (and absent some form of fraud), we sleep well at night because we believe a company’s management should be quite “willing to pay” lest they risk wiping out shareholders and losing their own jobs in the meantime.
Emerging market government debt also involves an assessment of both capacity and willingness to pay, although in this case it is the capacity and willingness of a country, not a business, and it’s our sense that few investors have any real skillset in assessing either one. An evaluation of each might start with something along the lines of understanding the mindset of the electorate, to which we can contribute absolutely zero. It is the elected government that will ultimately make budgeting decisions, determine taxation policy, regulate business and oversee the shape and size of public and private sectors. Then politicians will decide whether they have any interest in paying off debt owed to foreign investors. Let’s review some case studies.
Don't cry for me, Argentina
Argentina is an interesting test case on a number of fronts. This is a country that has managed to default on its debt no less than six times in the 72 years since the end of WWII: 1951, 1956, 1982, 1989, 2001, and 2014 – once every 12 years on average over the periodv. Its various governments have exhibited little interest in running the country in a way that ensures capacity to pay its obligations, nor much willingness to do so when it can. And yet, in June 2017, Argentina issued a 100-year bond denominated in U.S. dollars. Interest payments of 7.185% are scheduled to be paid annually until the principal is repaid in the year 2117 (seeing that principal returned is what today’s investor might refer to as “someone else’s problem”)vi.
Still, those 100-year bonds were bid up to a price of US$105 to start the year, only to fall to US$70 by the end of August on a proliferating bribery scandal and the Central Bank raising its overnight rate to 60% to shore up a currency that’s declined by 55% year-to-date against the U.S. dollarvii. It’s Déjà vu all over again.
Republic of Argentina 100-year 7.125% bond
Dec. 31, 2017 to Sep. 30, 2018
Source: Bloomberg LP. As at September 30, 2018. Price in US$.
Indeed, even if you could wrap your head around the strength and resilience of a country’s economy and the will of its politicians, weakening currencies and declining exchange rates can also pose a conundrum. Argentina wasn’t the only emerging market economy to take advantage of accommodating US$-denominated debt markets over the past decade. Shifting exchange rates mean that the problems that arise as tax dollars raised in weakening currencies are required to pay obligations denominated in a strengthening U.S. dollar are beginning to come to a fore. The chart below highlights the change in US$-denominated debt as a percent of gross domestic product (GDP) for a number of emerging market issuers from 2008 to 2017.
Emerging market exposure to US$ appreciation
US$-denominated credit to non-bank borrowers, % of GDP
Source: Rangasamy, Krishen. "Hot Charts - Economics and Strategy". National Bank of Canada Financial Markets. August 14, 2018 (Vol. XIX, No. 71). https://www.nbc.ca/content/dam/bnc/en/rates-and-analysis/economic-analysis/hot-charts-180814.pdf. Accessed on October 1, 2018
Like Argentina, Turkey’s currency has experienced a downward spiral. The chart below shows the year-to-date change in the Turkish lira-to-U.S. dollar exchange rate. At the start of the year, it cost 3.80 lira to pay off every US$1.00 of debt compared to more than 6.05 lira at the end of Septemberviii. According to the chart above, all things being equal, the appreciation of the U.S. dollar against the Turkish lira increased US$-denominated debt outstanding from 23% of GDP to over 38% in just eight months. If we aren’t any good at predicting the capacity and willingness of some future political party to pay off its debt, we’re even worse at forecasting the direction of its currency.
Turkish lira to U.S. dollar exchange rate
Dec. 31, 2017 to Sep. 30, 2018
Source: Bloomberg LP. As at September 30, 2018.
Government bonds across the board are unattractive investments in our view, and few investors have the skillset required to add tremendous value in the space. Government bonds in Canada, the U.S. and other developed economies have been a dismal place to invest in recent years and prospects haven’t improved. We can appreciate the idea of including government bonds in a diversified portfolio given the historical relationship that has seen them rise in times where stocks have declined, but that characteristic might not be so dependable in an environment where rising interest rates weigh on asset prices more broadly. Still, the unattractiveness of domestic government bonds doesn’t mean we need to venture to emerging markets for the premium they might offer. We will continue to look to create meaningful value for our investors by assessing businesses and taking advantage of the dislocations between the rate of return on their debt in relation to the risks of the enterprise. For that we will need to understand future profitability, not the next political party.
iiSource: Bloomberg LP. Bloomberg Barclays US Treasury Total Return Index weighting as at September 30, 2018.
iiiSource: Bloomberg LP. As at September 30, 2018
ivSource: Bloomberg LP. FTSE 10-year Treasury Benchmark On-The-Run Index and US CPI Urban Consumers Less Food & Energy SA used to calculate annualized return numbers and dollar values.
vRogoff, Kenneth. “Argentina is not solely to blame for its latest debt default.” TheGuardian.com, August 1, 2014. https://www.theguardian.com/business/2014/aug/01/argentina-blame-debt-default. Accessed on October 1, 2018.
viMander, Benedict & Wigglesworth, Robin. “How did Argentina pull of a 100-year bond sale?” Financial Times, June 20, 2017. https://www.ft.com/content/5ac33abc-551b-11e7-9fed-c19e2700005f. Accessed on October 1, 2018.
viiSource: Bloomberg LP. As at September 30, 2018.