Can We Predict Bond Returns? (04/02/2015)

One of my favorite sources of weekly analysis are “The Weekly View” pieces from RiverFront Investment Group. Their team came up with a number of really cool analytical frameworks and one of them is shown in Exhibit 1.  The basic idea is that the starting yield on a 10-Year Treasury bond is a very good predictor of its total return over the subsequent 10 years.  So if the yield today is 1.9% (as of 4/2/15) then we can expect our annualized return to be 1.9%, give or take a few basis points.

Exhibit 1 – RiverFront’s Bond Return Framework

Intrigued by such a simple model, I decided to run my own numbers to see how it works and if I can glean any additional insights. The results of my admittedly more amateurish analysis are shown in Exhibit 2. The 10 Years panel recreates RiverFront’s chart with annual instead of monthly data. It shows strong correlation between starting yield and returns: R-square of 0.9 is quite high and there is a clear linear relationship. The next three panels drop the projection timeframes to 5, 3 and finally 1 year. As you can see in the correlation equations, the accuracy of return predictions declines for shorter periods. Visually, the datapoints exhibit more dispersion with the 1-Year panel spread all over the place.

Exhibit 2 – Forward Annual Returns Based on Starting Yield

Source: FRED, Robert J. Shiller, Aswath Damodaran, PlanByNumbers

So what does it all mean for people buying bonds (or bond funds) today? The equations suggest that bond investors should expect to make roughly 1.6% over the next 10 years (Exhibit 3). Keep in mind that those are nominal returns before inflation, which could easily be higher than that and result in negative real returns. Five year projection is a little higher at 2.3% but comes with lower “accuracy” of 0.8.  I wouldn’t pay much attention to the shorter timeframes, especially the 1-year randomness.

Exhibit 3 – Projected Forward Returns on 10-Year Treasuries

Overall, I would take these return projections with a grain of salt. One general takeaway is that bond investors should temper their expectations, which is not easy to do after a 30-year bull market.

2014 Review: Economy Edition (02/04/2015)

U.S. GDP numbers were released last Friday, which means we can now finish 2014 year-in-review series by taking a look at major economic indicators. The employment situation continued to improve in 2014 (Exhibit 1).  The U.S. economy added 3.47 million new jobs (289,000 a month), which was a 60% improvement over 2013.  Total employment increased 2.5% which was quite a bit better than the 0.6% population growth.  The unemployment rate ended the year at 5.4%, which is in the range of what Fed currently considers full employment (see footnote).  U-6 rate is a broader measure defined as “Total unemployed, plus all marginally attached workers plus total employed part time for economic reasons”.

Exhibit 1 – Employment

GDP growth remained in the 2% range last year (Exhibit 2).  Auto sales jumped 8.9% to over 17 million a year (the highest level since 2005).  Inflation, as measured by Consumer Price Index, actually decreased and came in at 0.8%.  This level is well below the Fed’s 2% target and something to keep an eye on as everyone is trying to figure out when it will start raising rates.

Exhibit 2 – Growth & Inflation

Public debt increased to \$17.8 trillion during the year (Exhibit 3).  On the positive side, a combination of tax hikes and spending cuts led to a big drop in budget deficit.  It declined by 30% to \$483 billion or (only) 2.8% of GDP.  Despite the “Taper” , Federal Reserve still expanded its balance sheet to \$4.2 trillion.  Average monthly increase was about \$40 billion (down from \$90 billion in 2013).  It should actually decline this year as securities mature and there are no fresh purchases. We’ll do a separate post analyzing Fed balance sheet in more details.

Exhibit 3 – Debt & Deficit

Note: Public debt figure as of Q3:2014

S&P 500 earnings growth decelerated to 7.9% (Exhibit 4).  P/E multiple expanded 3.3% to 17.8x.  10-Year Treasury rate declined to 2.2%, which was totally unexpected as most market pundits predicted a big jump in rates for 2014. 30-Year Fixed Mortgage rates also declined ending the year just under 4%.

Exhibit 4 – Earnings & Rates

Note: Q4:2014 S&P 500 earnings are consensus estimates as of Jan 22, 2015

Finally, the housing sector continued improving (Exhibit 5).  Both units and prices increased again in 2014, albeit at a much slower pace than in recent years.

Exhibit 5 – Housing

Note: S&P Case-Shiller 20-City Home Price Index as of Nov 2014

Footnote

Federal Reserve currently considers full employment to be in the 5.2% – 5.5% range.

Committee participants’ estimates of the longer-run normal rate of unemployment had a central tendency of 5.2  to 5.5 percent.

http://www.federalreserve.gov/faqs/money_12848.htm

More traditionally “Full Employment” in the U.S. means 4%.

The United States is, as a statutory matter, committed to full employment (defined as 3% unemployment for persons aged 20 and older, 4% for persons aged 16 and over); the government is empowered to effect this goal. The relevant legislation is the Employment Act (1946), initially the “Full Employment Act,” later amended in the Full Employment and Balanced Growth Act (1978).

http://en.wikipedia.org/wiki/Full_employment

Economy in Perspective, an 85-Year One

Doing a post on the economy in 2013 got me thinking about how current environment compares to prior decades.  I tried to focus on a very small number of indicators that are truly representative of the national situation and not get bogged down in a forest of minute details.  The numbers in the tables and charts below are averages of the annual data for each decade.

Exhibit 1 presents eight indicators with the highest number for each row highlighted in green, while the lowest is highlighted in orange.  Not surprisingly, the 1930s stand out quite a bit as crappy time to be around.  If we make an argument that 1930s and 40s were “anomalies” what with the Great Depression and World War II, things look a bit different (Exhibit 2).  The 1950s was a model boom decade with a roaring stock market, low interest rates, fast-growing economy, relatively low inflation and a population “baby boom”.  Both the unemployment rate and budget deficit were very low, while the high WWII debt was being paid off.

On the flip side, the 2010s have been pretty crappy so far as indicated by the dominance of highlighted numbers in that column.  Attractive stock market returns are the major exception here.  Of course, we are only 4 years into the decade and things might look quite a bit better down the road.

Exhibit 1 – Important Metrics by Decade (1930s to 2010s)

Exhibit 2 – Important Metrics by Decade (1950s to 2010s)

Now let’s take a look at these metrics one by one in a bar chart form.  These are averages of the annual data for each decade.  Stock market returns have been remarkably attractive for a majority of the 20th century decades, save for 1930s and 2000s (Exhibit 3).  The negative returns for the 2000s certainly help to explain continued investors’ reluctance to invest in equities.  Although fund flows for 2013 show that this might be changing.  Interest rates look like a bell-shaped curve rising in mid-century and culminating in double digits in the 1980s.  From there we began the 30-year decline in rates with a coincident bull market in bonds.  What this chart looks like say 50 years from now is anybody’s guess, but there is not much room left to the downside in interest rates.

Exhibit 3 – Stock Market Returns & Interest Rates

World War II and the subsequent rebuilding led to an economic boom in mid-20th century (Exhibit 4).  Real GDP (adjusted for inflation) is still growing albeit at a slower pace.  Speaking of inflation, after peaking in 1970s (which led to high interest rates in Exhibit 3), it has been steadily declining.  In fact, despite massive “money printing” by the Fed as well as ballooning federal debt and budget deficit (Exhibit 6), inflation in this decade has been the lowest since 1930s.

Exhibit 4 – GDP Growth & Inflation

There is a reason the post-war period was nicknamed “baby boom” with strong population growth (Exhibit 5).  With a booming economy, there was plenty of work to go around leading to low unemployment rates.  Once more, 2010s have been a rather tough time to be looking for work.

Exhibit 5 – Population Growth & Unemployment Rate

Finally, the federal debt and budget deficit have seen major changes over the last few decades (Exhibit 6).  During World War II, the government borrowed heavily to support war effort, but slowly reduced the debt burden into 1970s.  The 1990s and 2000s saw a relatively stable debt load, but it increased markedly after the “Great Recession”.  Similarly, current budget deficits are quite high by historical standards (although last few years have trended down).

Exhibit 6 – Federal Debt and Budget Deficits

This piece is meant to put current environment into historical perspective and is something I will likely refer to quite a bit in future posts as well as my work with clients.