Submitted by Nicholas Colas of DataTrek
What will the S&P 500 do over the next 20 years? Unknowable, of course, but still analyzable through the lens of market history. The last 20 years have been the worst for compounded market returns since the Great Depression. The next 20 years can be better, as long as inflation remains low and (more importantly) technology improves workforce productivity. Every big cycle (the 1950s/1960s and 1980s/1990s) has its “thing” – technology has to be that driver now.
* * *
Last night we outlined how the S&P 500 and the Technology sector writ large are essentially tied at the hip. Tech plus Google, Facebook and Amazon are 29% of the S&P. That’s a function of Tech’s dramatic outperformance over the last 5 years, but we posited that it leaves the next 5 years of US stock performance equally beholden to Tech’s fortunes.
That got us thinking about a topic we last discussed in late 2018: future long run returns on the S&P 500. Our definition of “long run” is 20 years, both because that is indisputably a good multi-cycle time frame and the historical S&P 500 return data shows very distinct patterns using 20-year compounded annual returns.
Looking at trailing 20-year periods back to 1928, there are five distinct periods for US stocks:
#1: The Great Depression into World War II. Starting points matter to long run returns, and 1928 was a really bad one. A dollar invested on December 31st of that year didn’t get back to breakeven until 1936. And if you didn’t sell then, you were sitting on a loss again until 1944. All that made for very poor long run compounded returns:
- From 1948 – 1951 trailing 20-year compounded returns on the S&P ranged from 2.4% to 6.6%.
- Inflation-adjusted returns were just 0.6% - 4.3%.
#2: World War II and into the 1960s. The S&P 500 doubled during the war. The all-time best year for US stocks was 1954, with a 52.5% total return. And with no serious drawdowns over this time span, compounded returns were very good:
- 20-year trailing compounded returns rose consistently from 1952 (meaning a starting point of 1933) to 1962 (starting in 1943).
- The high water mark for trailing returns was in 1962, at a 16.7% compounding rate over the prior 20 years. In layman’s terms means an investment in the S&P doubled every 4.3 years across that two-decade span.
- Inflation adjusted returns were also strong, at 13.3% from 1943 – 1962, but pick any end point from 1960 to 1970 and the prior 20 years all posted compounded growth rates of +10% after inflation.
#3: High inflation/high interest rate 1970s. The 1973 oil shock along with higher interest rates required to tame inflation put an end to double-digit compounding rates for the S&P 500. Long run returns, especially after considering inflation, suffered badly:
- US stocks dropped by 38% in 1973/1974, and even after 2 better years in 1975 and 1976, they were unchanged from 1972 when 1977 came to a close.
- That was enough to push 20-year compounded returns down to 6.5% for the 2 decades ending in 1979, and inflation-adjusted compounded returns were just 1.0% for the 20 years from 1961 – 1980.
- Compare these returns to the prior period. Twenty year trailing nominal returns went from 16.7% (1962 end point) to 6.5% (1979 end point). Inflation adjusted returns went from 13.3% to basically zero (1.0%).
#4: Lower inflation, lower rates, dot com boom. From 1980 to 1999 the S&P was only lower in 2 years: 1981 (-4.7%) and 1990 (-3.1%). No surprise therefore that this was the best period for US stocks since, well, ever:
- From 1980 to 1999 the S&P 500 compounded at 17.6%, doubling every 48 months on average. Inflation adjusted compounded returns were 13.1%.
- Again, consider how different this cycle was from the one in point #3: long run returns of 17.6% here are more than double the 6.5% returns then. And inflation-adjusted returns went from 1.0% to 13.1%.
#5: The current volatility – prone market. You know what’s happened since 2000 in terms of returns: the S&P 500 has lost a third of its value twice (2000-2002, 2008) in the last 20 years. That has crushed long-term returns:
- For the 20 years ending 2018 the S&P 500 has compounded at a nominal 5.5% and an inflation-adjusted 3.0%.
- That is the lowest 20-year CAGR since the data from the post-Great Depression data in point #1.
As far as what this history says about the next 20 years for the S&P 500, three closing thoughts:
- Financial stability matters. Nothing kills long-term returns quicker than a large drawdown from an equity market bubble (1929 - 1931, 2000-2002) or systemic excess (2008). Will we have a financial crisis/market bubble burst over the next 20 years? Recent history says “of course”, with the counterbalance being a larger monetary/fiscal policy playbook to offset its effects.
- Low inflation really helps. The 1970s lost decade came from 2 oil shocks that revealed the fragility of US monetary policy. Now, we have low goods-and-services inflation around the world, and with an aging global population inflation should remain at bay.
- Big returns need an equally big catalyst. Peak long run returns in the 1960s and 1990s came from large macro themes: US growth post WWII and declining interest rates/Internet adoption.
That’s a good place to end because it’s where we started this note: America’s development and use of technology will be the key determinant of long run future S&P 500 returns. With population growth less than 1%, economic growth will have to come from productivity. And it’s not just Technology companies that will have to benefit from this trend. Whatever they develop will have to make every other industry more efficient and scale to the rest of the world.
Source for return data: NYU Professor Aswath Damodaran: