An Update On The Bond MarketMay 10, 2021
- Long-term bonds have been a powerful addition to any portfolio over the last few decades. Over the multi-year view, that will remain the case.
- Long-term bonds perform poorly when nominal GDP growth is increasing or is expected to increase. This is because the long-term Treasury rate will move with the long-term fluctuations in nominal GDP growth.
- High levels of government involvement in the economy, mainly in the form of transfer payments, will reduce long-term growth.
- Weaker levels of long-term growth will create excess capacity and a continuation of the persistent disinflationary trend. In the short term, however, the cyclical upturn in growth is still ongoing.
- I will wait for the leading indicators of growth and inflation to start a new downturn before adding an overweight position in LT bonds.
An Update On The Bond Market
(To expand images, you can "right-click" and select "open link in new tab.")
Long-term Treasury bonds have been in a multi-decade bull market as nominal GDP growth has decreased over time.
The biggest determinant in the long-term Treasury rate is the direction of nominal GDP growth. Treasury bonds have two major components: The real interest rate and inflation expectations. Nominal GDP growth also has two components, real growth and inflation. Thus, nominal GDP growth and the long-term trend is the most influential factor in determining the direction that long-term bonds yields will move.
The persistent decline in long-term bond yields has puzzled many investors, particularly as the level of indebtedness in the government sector has increased, the Federal Reserve balance sheet has expanded, and fiscal policy has become more extreme.
If we map the long-term (20-year) rate of growth in nominal GDP, the decline in bond yields becomes much more clear.
Nominal GDP growth has fallen from a long-term trend of 8% to just 3% over the last 20 years.
Nominal GDP Growth: 20-Year Annualized Rate
Source: BEA | To Expand, Right-Click > "Open Link In New Tab"
When analyzing nominal GDP growth, both the secular trend and the cyclical trend are critical. The secular trend is the slow-moving, multi-year change in the rate of nominal GDP growth. The secular trend is impacted by demographics, productivity, debt, and more. The secular trend offers guidance on the 3-5 year view on long-term Treasury bonds and like securities such as long-term bond ETF (TLT).
There are cyclical movements within a long-term secular trend, generally lasting 6-18 months on average, but there's room for potentially longer cyclical trends.
Any period of time shorter than a couple of quarters is far too short for any help from economics or cyclical analysis.
Cyclical trends in nominal GDP growth can move counter to the long-term trend. For example, since the summer of 2020, the US and global economy have been in a cyclical upturn, creating upward pressure on interest rates and downward pressure on bond prices, despite the longer-term secular trend that still argues for lower growth and lower interest rates.
In the sections below, I will briefly update the secular and cyclical view for nominal GDP growth and, thus, long-term Treasury bonds.
This framework has produced strong results. In October 2018, when 30-year Treasury rates were heading near 3.5%, I offered a bullish analysis on bonds and a forecast for much lower interest rates.
Growth is going to surprise to the downside over the next two quarters... The early indicators of the economy have moved notably lower.
Buying a 3.35% 30-year Treasury with the potential for new secular lows in interest rates over the next several years has enormous profit potential.
I am still a buyer of the long bond.
"Should You Be Worried About This Rise In Interest Rates?" - October 2018
As the economy started to show signs of acceleration based on the cyclical leading indicators, it was time to lighten up on Treasury bonds and start adding commodity exposure to hedge an inflationary impulse. Therefore, in February, I reiterated the cyclical view based on the leading indicators of growth and inflation:
The economy remains in a cyclical upturn which should result in higher bond yields... The focus should remain on the economy's cyclical indicators, which suggest the preference should remain for stocks and commodities over bonds and gold, relative to your chosen baseline or benchmark allocation.
"Taper Tantrum 2.0" - February 2021
The long-term structural outlook will provide a solid framework and keep your portfolio on the same side as the gravitational economic force.
Regular updates to the economy's cyclical leading indicators will provide clues about when it's time to overweight long-term bonds as a recession approaches or lighten up on longer-duration bonds when interest rates are subject to rise.
As mentioned earlier, the long-term trend in economic growth is primarily a function of demographics (population growth) and productivity.
While the demographics for the United States are better than Japan or Europe, population growth has still been in a declining trend.
For example, in the 1950s, the United States had population growth as high as 2.2% compared to 0.6% today. By 2030, US population growth is expected to decline to a rate below 0.5%, so demographics will be a continual drag on economic growth. This factor alone will keep a lid or a ceiling on bond yields over the multi-year view.
The second factor, productivity, can be proxied via a metric called the marginal revenue product of debt or MRPD. The MRPD calculates how much growth the economy is generating per one dollar of debt.
Good uses of debt will increase this metric, while bad (unproductive) uses of debt will lower this ratio.
In the 1960s, the US was generating over 80 cents of growth per dollar of debt. Today, that metric has fallen to roughly 35 cents.
Adding these two factors together, the trend in population growth and the MRPD implies the US will sink to a long-term trend of roughly 1% real growth and even lower a decade in the future.
If real economic growth trends lower, the economy will have an abundance of resources, including too much labor, which will weigh on wage growth and overall price inflation over the 3-5 year view.
MRPD + Population:
Source: Our World In Data, Federal Reserve, BEA | To Expand, Right-Click > "Open Link In New Tab"
We can arrive at the 1% growth trend another way.
Economists Andreas Bergh and Magnus Henrekson studied government involvement in an economy and concluded that as government size or government intervention increases by 10 percentage points, an economy loses 0.5 to 1.0 percentage points of annual growth.
The best measure for government involvement, aside from debt to GDP ratios, is the percentage of total income that comes from government transfer payments.
Government transfer payments is a broad category, including social security, unemployment, stimulus payments, and more. For lack of a better term, this ratio shows what percentage of total income is distributed or "re-distributed" by the government.
In the 1960s, the government was distributing around 5% of total income in the country.
Before the pandemic, that percentage had increased to roughly 17%.
Total Transfer Payments As A % of Total Income:
Source: FRED | To Expand, Right-Click > "Open Link In New Tab"
The chart below shows the annualized growth rate in real GDP per capita for each expansion ending in the year in the chart. The last expansion, ending in 2019, only had a growth rate of 1.66% in real per capita terms, over one percentage point lower than the period from the 1970s to the 1990s.
Change in Real GDP / Capita By Expansion:
Source: FRED, BEA | To Expand, Right-Click > "Open Link In New Tab"
As a result of COVID and the major stimulus bills, the government is currently distributing almost 35% of total income.
The ratio will not stay at 35%, most likely, but as a country, we will decide on a new level of transfer payments, likely higher than the pre-COVID baseline of 17%.
If we settle on a new level of roughly 25%, up from 17%, we can expect to lose more growth relative to our pre-COVID trend. If we sustainably increase government involvement in the economy by roughly 10 percentage points, we can expect to lose another 0.5 to 1.0 points of growth, placing real GDP per capita around the 1% level for the decade ahead.
If the government controlled 100% of the income distribution in the country, most people would agree the economic results would be worse than a free market.
One way to look at the chart above is that the US economy has been transitioning from a roughly 95% free economy to now a 65% free-market economy as government involvement has increased.
As the economy moves closer and closer toward a command and control economy, the results measured by real GDP per capita will deteriorate.
If real GDP per capita growth declines further, it will create an abundance of resources that will weigh on the overall rate of inflation.
The secular trends, analyzed in many ways, all point toward a decade of weaker growth, likely sub 1% after the economy fully experiences the pent-up demand rebound that is common after deep recessions.
Lower trend levels of growth will create lower long-run inflation. The combination of weaker growth and lower inflation means that nominal GDP growth will stay in a downward trend, reducing the long-term interest rates over the multi-year view.
While the long-term secular trends are decidedly lower, the economy is still in the middle of a strong cyclical upturn that started in the summer of 2020.
A pent-up demand rebound in economic activity and productivity is typical of a post-recession year.
This current cyclical upturn in economic growth has been exacerbated by a change in consumer behavior, shifting away from service consumption and toward durable goods consumption.
This shift created an imbalance in the global manufacturing sector at the same time that supply chains were damaged from the COVID pandemic.
As a result, bottlenecks emerged, and prices for raw materials surged.
We can see the powerful cyclical trend in some of the sub-components to the ISM manufacturing report, including the prices paid index, supplier delivery times, and a self-constructed index of new orders to inventory.
Source: ISM | To Expand, Right-Click > "Open Link In New Tab"
The CRB index of raw industrial commodities has increased at an 83% annualized pace over the last three months, which signals the bottlenecks and demand for raw materials is still strong. As a result of the ISM sub-components, the rapid increase in raw industrial commodity prices, and the host of other cyclical leading indicators we track at EPB Macro Research, the risk of a new cyclical downturn in the next couple of months is very low.
CRB Raw Industrials: 3-Month Annualized Growth
Source: CRB | To Expand, Right-Click > "Open Link In New Tab"
When the cyclical trends in the economy are higher, stocks and commodities generally outperform bonds and gold.
Despite the powerful secular trends, I continue to hold an underweight allocation to long-term Treasury bonds relative to my personal baseline portfolio.
I plan to shift back to a heavy overweight stance on long-duration Treasury bonds when the leading cyclical indicators of the economy turn lower.
Summary and Outlook
The most powerful long-term factors shaping the secular trend in economic growth, namely demographics and productivity, or population and debt, are decidedly lower.
Over the next several years, the economy will struggle to grow at a faster rate than the depressed 1.7% real per capita growth trend of the last expansion ending in 2019.
The increased level of government involvement will expectedly reduce trend growth, consistent with the analysis on population and the MRPD.
Cyclically, however, the economy's most cyclical leading indicators have not yet turned lower, which means a soft patch for long-term bonds is still on the horizon.
I hold a core position in long-term bonds as a part of my balanced all-weather asset allocation framework.
I plan to move from my current underweight stance on long-term bonds back to an overweight position when the cyclical leading indicators turn lower.
If you are looking for regular updates on the short-term cyclical trends in the economy, as well as asset class preferences, consider subscribing to our monthly Cyclical Leading Indicators report.
One time per month, for an extremely low price of only $12.99/mo, we outline the most important cyclical leading indicators of the economy and review asset class preferences in the context of equities, fixed income, precious metals, and commodities.
For more information on the EPB Cyclical Leading Indicators product, click here.
We also publish a quarterly presentation covering both long-term (secular) economic trends as well as short-term (cyclical) economic trends and the resulting impact on asset prices such as equities, fixed income, gold, and more.
For more information on the EPB Quarterly Presentation & Outlook product, click here.
Join The Free EPB Macro Research Blog
New research, content, and special offers delivered straight to your inbox.
We hate SPAM. We will never sell your information, for any reason.