Overview: America is riding on a slow moving turtle
Over the last two years Lawrence Summers has been an energetic proponent of the secular stagnation thesis. He argues that secular stagnation has occurred because of a chronic excess of savings over investment, an imbalance currently present in the U.S. and
other developed markets. This framework posits that the increased propensity to save and the decreased predilection to invest acts as a drag on demand, reducing both growth and inflation, and pulling down real interest rates.
According to the secular stagnation thesis, it is difficult for an economy thus afflicted to exhibit significant growth. As Summers emphasizes, when such growth is achieved—as in the U.S. in the late-90s or between 2003 and 2007—it comes from dangerous levels of borrowing that translate excess savings into unsustainable levels of investment (in these two instances, resulting in tech and housing bubbles).
Explanations for the decline in GDP growth and interest rates that are complementary to secular stagnation include:
• Kenneth Rogoff’s debt overhang argument,
• Robert Gordon’s supply-side headwinds (weak trend productivity growth) explanation and
• Ben Bernanke’s savings glut commentary.
Although secular stagnation offers the most comprehensive account of the current economic situation, it can be viewed as subsuming Rogoff’s, Gordon’s and Bernanke’s explanations.1 The main components of this framework are now discussed.
The reduction in debt levels (Figure 1) implies more saving and less consumption by households, and relatively tight fiscal policy by governments. Further, this is an issue for almost all G10 countries, as well as many emerging markets (notably China, where
corporate debt levels are of particular concern).
Figure 1: U.S. debt ratios
U.S. debt ratios are elevated for both households and the federal government.
The U.S. labor force will grow at a substantially lower rate over the next two decades than it has since 1950 (Figure 2). This is another growth headwind that is shared with most other developed markets. The point is further reinforced if we use a quality-adjusted labor force metric, in recognition that improvements in education levels are plateauing in the U.S. and other OECD countries. Additionally, population growth is even lower in most large countries (e.g., China, Japan, Germany, France, Spain, Italy, and U.K.).
Figure 2: U.S. population growth rate
The U.S. has experienced a significant decline in population growth.
3. Rising inequality
There have been two key changes in the distribution of income: first, between labor income and capital income (Figure 3); and second, between those individuals with more wealth and those with less (Figures 4–6). Both trends have raised the propensity to save,
creating yet another headwind for consumption and overall GDP growth.
Figure 3: U.S. corporate profits
Profits remain high as a % of GDP
Figure 4: U.S. Gini index
The Gini index shows that U.S. income inequality has increased significantly
Figure 5: U.S. income distribution
A rising share of U.S. income is received by top earners, who have a lower propensity to consume
Figure 6: U.S. savings by income category
Top income quintiles in the U.S. have higher propensities to save (and correspondingly, lower propensities to consume)
Source: Dyman et al, Federal Reserve Board, “Do the rich save more?”
4. Capital-light world and the declining price of capital goods
There has been a reduction in the investment intensity of corporations, reflecting changes in the structure of the economy (Figures 7–9). In particular, manufacturing has experienced a declining weight, while the less capex-intensive service and technology sectors have become more important.
Additionally, cheaper capital goods (Figure 10) means that a targeted level of investment can be achieved with less borrowing and spending, again reducing the propensity for investment. To illustrate, the lower price of capital goods, especially information and communications technology (ICT), has driven a declining trend in the sector’s capex/sales ratio over the last 25 years.
Figure 7: U.S. capital stock
The growth rate of the U.S. capital stock has declined precipitously
Source: Penn World Tables
Figure 8: U.S. net business investment vs. corporate profits
The percentage of U.S. companies with zero inventories is increasing
Figure 9: U.S. companies with zero inventory
The price of U.S. capital goods has declined dramatically relative to consumer goods
Source: ClariFi; Morgan Stanley. Note: Excludes impact of tech bubble.
Figure 10: U.S. capital goods versus consumer goods
The price of U.S. capital goods has declined dramatically relative to consumer goods
Source: Bloomberg. Using the producer price index for capital goods and the CPI for consumer goods.
5. Slow productivity growth
Summers is agnostic on this issue (“there is the possibility, on which I take no stand, that the rate of technological progress has slowed”), but Gordon has written volumes on it. He states that innovations unleashed from 1870 to 1970 were exceptional, and unlikely to be repeated. Although he is not suggesting that the pace of innovation will slow further, he believes it will be in-line with the post-1970 experience (Figures 11 and 12).
This perspective is contested by the techno-optimists (Brynjolfsson, McAfee and others) who predict that unprecedented breakthroughs in technology (e.g., AI, genetics, 3D printing), will drive a faster pace of technological change over the next 25 years. Gordon believes their arguments are hyped and unconvincing. That said, many past advances came as complete surprises and few people appear capable of accurately predicting trends in innovation and productivity.
As an additional point, Gordon agrees that national accounts statistics understate the true extent of progress, but he emphasizes this has always been the case and was probably a more important factor during the 1870-1970 transformation than it is today.
Regardless, Gordon and Summers reach similar conclusions regarding the outlook for GDP growth and real interest rates, but Gordon emphasizes supply-side issues, whereas the focus of Summers is on the demand-side.
Figure 11: U.S. total factor productivity (TFP) growth
TFP averaged over 1.5% for the five mid-century decades. However, for the rest of the time, 0.7% is about normal.
Source: Gordon 2014 and San Francisco Federal Reserve. Note: Averages for preceding decade (except 2015, which is 2010–2015)
Figure 12: U.S. productivity growth
Productivity in the U.S. has remained tepid since 2010
6. Chinese rebalancing and deleveraging
Similar to slowing productivity, this factor did not play a role in earlier versions of the secular stagnation thesis. However, it is probably more important than many pundits acknowledge, given that Chinese growth is likely to continue slowing and that it drove over 40% of post-2009 global growth. Former Reserve Bank of India Governor Rajan believes China’s excess capacity and high leverage is a key element of secular stagnation.
We agree with Governor Rajan’s opinion and conclude that Chinese deleveraging and rebalancing has several important consequences:
– Excess capacity has resulted in China exporting goods-price deflation to the rest of the world.
– Less investment in China has driven lower commodity prices (e.g., iron ore, steel, coal and copper).
– China’s ongoing deceleration poses a strong headwind for global cyclical sectors.
– Chinese monetary policy will need to loosen even further, which would drive a weaker RMB and a lower and flatter yield curve.
Figure 13: Composition of china’s GDP
Chinese growth is rotating away from investment towards consumption
Figure 14: Chinese industrial Production
Chinese industrial production and investment growth continue to decelerate, although there has been an uptick in the Premier Li index (bank loans, electricity consumption, rail freight).
Secular stagnation and the 400+ bps decline in real interest rates
Real interest rates in the U.S. have ratcheted down with each recession over the last few decades (Figures 15 and 16). A similar trend has occurred in all other major government bond markets, which is one reason why central banks are so worried about their ability to respond to the next recession.
Figure 15: U.S. real bond yields
Yields have declined precipitously since the early 1980s (a comparable move occurred in all major markets), with the lion’s share due to secular stagnation factors.
Figure 16: U.S. real Federal Funds Rate
The Federal Funds Rate has followed a similar downward trajectory
To understand this move two economists from the Bank of England, Rachel and Smith (2015), analyzed the 450 bps decline in global real interest rates over the past 30 years. They examined three factors that affected savings and three that impacted investment decisions, and were able to explain 400 bps of the 450 bps decline.
Rachel and Smith demonstrated that savings had increased for three reasons:
– Demographics — an aging population saves more, which was estimated to reduce interest rates by 90 bps.
– Inequality — wealthy people have a lower propensity to consume, which was found to be responsible for 45 bps.
– EM precautionary savings — which followed the Asian crisis accounted for 25bps.
Similarly, they showed that the investment curve has shifted down for three reasons:
– The price of capital goods has declined dramatically since 1980 (especially ICT); estimated to have reduced interest rates by 50 bps.
– Public investment has fallen, particularly for developed markets and may continue to fall with the imperative of fiscal tightening responsible for 20 bps.
– Return on capital has declined by less than the risk-free rate, implying a higher investment premium, accounting for 70 bps.2
– Further, 100 bps is attributed to the deterioration in trend growth, leaving only 50 bps of the 450 bps decline in global real rates unexplained by Rachel and Smith’s analysis.
A consensus is developing among economists and policy makers that unorthodox monetary efforts are becoming less and less effective, while greater emphasis is likely to be placed on fiscal spending (infrastructure, as well as education, job training, research and development, and other programs to improve growth and increase competitiveness).
Critics of monetary policy employ liquidity-trap and Zero Lower Bound type arguments of varying degrees of complexity and sophistication. Fiscal policy acolytes note that: borrowing rates are extremely favourable; most economies are operating with significant slack so little activity would be crowded out; and infrastructure spending has been neglected in many developed economies (Figures 17 and 18). All this suggests compelling fiscal multipliers, although the IMF’s research emphasizes the need for a tailored approach across and within countries.
Figure 17: U.S. total public construction spending
Growth averaged 6.7% from 1965 to 2007, but only -0.6% from 2008 to 2016
Several countries have already announced fiscal stimulus plans, notably Japan and Canada, with China always in the foreground. Additionally, we expect fiscal policy loosening to be announced in the U.K. next quarter and in the U.S. from the second quarter of 2017 (regardless of who wins the presidency).
Figure 18: Japanese real public investment
Japanese public investment has also fallen sharply
Further, many Eurozone countries would like to loosen their fiscal stances (Italy has been the most outspoken), but are constrained by ideologues in Frankfurt, Berlin and Brussels. This means the ECB needs to carry the full burden and will almost certainly need to loosen policy even further. It also implies that Eurozone growth will continue to disappoint as policy makers squabble and stumble from one crisis to the next.
– Fixed income: The uber-bears are wrong, but there’s not much upside to holding bonds either.
– Commodities: A V-shaped recovery is unlikely given the tepid global growth outlook and headwinds facing investment spending.
– Equities: Tough environment for top-line growth and multiples are already stretched. Consequently, we prefer companies with strong track records regarding cash flow generation, shareholder distributions and capital allocation policies.
– Key downside risk: Wage growth in the U.S. accelerates, forcing the Fed to adopt a more hawkish tone, driving the yield curve higher and leading to a dislocative re-rating of the equity market.