VoxEU Column Financial Markets Labour Markets

Labour power sets the neutral real rate

The recent remarkably low interest rates have puzzled economists. The standard explanation rests on the extraordinary manoeuvres of the world’s largest central banks. This column argues, however, that it is due to economic developments, specifically globalisation and the collapse in labour power in the west.

The Bank of International Settlements has been among the most prominent of the many critics warning that central banks are inflating asset prices irresponsibly and in so doing, set the scene for financial instability.1 But the lack of inflationary boom associated with ultra-low rates in the west has prompted policymakers, academics, and industry figures to ponder whether we have entered an era of secular stagnation (Teulings and Baldwin 2014).

By contrast, I will argue that economic growth has been strong, and that the sustained fall in interest rates and rise in asset prices has been driven not by central bank policy but rather by economic developments. To connect these statements we need to consider the way that the world has globalised over the past 30 years. Asset prices are a function of the size of future cash flows attached to an asset, and the discount rates attached to these cash flows. Let’s consider each in turn.

Cash flows

Global economic growth has been stronger in the period since 2000 than in the two decades preceding it (Figure 1). However, economic growth of the advanced economies (AEs) has weakened over the same period. And this is despite – some would say because of – rising levels of debt and lower monetary policy rates that might serve to reduce future for present growth.2 We have though witnessed a change in the composition of global growth, which looks like a trend decline only from the perspective of the introspective advanced economy observer (Figure 2).

Figure 1. Global and advanced economy economic growth 1980-2015

Source: IMF, March 2015

Figure 2. Advanced and emerging economies share of global economy 1980-2015

Source: IMF, March 2015

But while the world economy has grown markedly and with this the cash flows attached to ‘real’ assets, the gains of globalisation have not been shared with uniformity. Figure 3 shows the distributional developments using household level data constructed by the World Bank. Four features are striking:

  • First, the biggest winners have been the ‘global middle’ which is largely accounted for by China’s meteoric rise;
  • Second, the very top of the income spectrum has done almost as well as the global middle;
  • Thirdly, there is a hard core of poverty that has not been touched by globalisation at the very bottom of the income spectrum; and
  • Four, and most interestingly, there is a large section of people who are well-off in global terms who have largely not participated in global growth over the past 20 years. That section is populated largely by the Western lower middle and working classes.

Figure 3. Change in real income 1988-2008 at various percentiles of global income distribution (2005 PPP USD)

Source: Milanovic (2012); Lakner and Milanovic (2013)

Western lower middle and working classes (what Marxist-Leninists would call the global labour aristocracy) have been left behind as a global labour market arbitrage/convergence trade has fuelled income developments at the top and the middle of the income distribution.3

This convergence trade can be seen clearly in Figure 4. It shows a rolling avalanche of labour supply over the 20 years to 2008 – and the extraordinary move out of poverty on the part of developing country populations, with the modal real income per household increasing from just under $400 per annum to just under $700 per annum. Under globalisation, location becomes less relevant to household income, while skills and endowments become more important.

Figure 4. Global distribution of income over time – logarithmic scale, population-weighted

Source: Milanovic (2012); Lakner and Milanovic (2013)

Marx’s Das Capital (1867) resonated with people in a way that it doesn’t today because class/ endowment largely determined your place in the global income hierarchy; location was less important.

Things have changed, and radically so (Figure 5). Disaggregating the Theil coefficient of inequality, we see that the main determinant of household income at the beginning of the 21st century is the country in which you were born. The huge ‘citizen’s premium’, enjoyed by citizens of rich countries, has been more recently challenged by the rise of the NICs, the fall of the Iron Curtain and advent of China to the global trading system. While the US has high Gini and Theil coefficients of inequality when compared to other developed countries, these are low in the global context.  Moreover, location plays a very small role in US household income. Things like inheritance (broadly defined), and the skills you acquire are much more important. If globalisation is to succeed, the world should look a lot more like the US. The cosmopolitanism of globalisation attacks the state’s communitarian raison d’être. So far this has resulted in an increase in both government debt to support welfare systems, and political heat around economic immigration. Rising income inequality in the West has furthermore fuelled interest in works such as Thomas Piketty’s Capital in the Twenty First Century – a work that takes an introspective Western perspective in which inequality is increasing rather than a global perspective in which inequality has begun to fall. It furthermore mistakes recent returns to asset-holders as proof of structurally high discount rates rather, as we will now see, as a function of falling discount rates.4

Figure 5. Level and composition of inequality in 1870 and 2000: world (disaggregated by countries) and the US (disaggregated by states)

Source: Milanovic (2012), Hisnanick and Rogers (2007), Columbia Threadneedle Investments March 2015.

Discount rates

The globalisation story to date has had two huge implications for asset markets.

  • The first implication is felt at firm level – how to respond to these changes. This is well known, so it will not be covered here.
  • The second concerns the impact on the real rate of interest. This is less well understood, but much more important for policymakers and investors.

As central bankers have frequently reminded us, the neutral real interest rate is set by the economy, not by the central bank.5 Globalisation and associated labour market convergence have been associated with a collapse in labour power in the west. When labour had power, the marginal costs of labour were high. As such, there was a big incentive to invest in capital to substitute labour for capital, and this brought the cost of capital higher. As labour lost power, wage pressures in the west collapsed.6 This led to a falling labour share of GDP, especially away from the top 20% of Western households (Figures 6 and 7).7 With lower labour costs, a reserve army of global workers whose size grew as trade barriers dropped and emerging countries developed, companies have increasingly been incentivised to substitute capital for labour, reducing the requirement for capital, and bringing the cost of capital lower in the west (Figures 8 and 9).8

Figures 6 and 7. Change in US (left panel) and UK (right panel) share of GDP, split by capital and labour income quintile, 1980-2010

Source: Bank of England, Congressional Budget Office, Office of National Statistics and Columbia Threadneedle Investments, March 2015.

Figure 8. US & UK bottom 80% labour share of GDP 1980-2010

Figure 9. Ten-year US Treasury and ten-year Gilt minus respective 12-month changes in Core Consumer Price Indices

Source: Columbia Threadneedle Investments and Bloomberg, March 2015.

The fall in the neutral real rate of interest (NRRI) in the west has been like in a textbook. It does not signify secular stagnation in economic growth rates, but instead the loss of labour power. And as the neutral rate has fallen, the present value of future promised cash flows has gone up in value. Figure 10 shows the total returns from investing in 15-year and 30-year zero-coupon US Treasury bonds (red and blue, respectively), and the S&P500 index of US stocks (green line). Over the past 35 years, assets have rallied largely in accordance with their sensitivity to changes in the neutral rate of interest:

The negative asset market implications of slow economic growth in the west have been less important than the positive asset market implications of weak labour pricing power.

Figure 10. 15-year, 30-year zero-coupon US Treasury bond total returns versus S&P 500 index total returns, 1982-2015

Source: Columbia Threadneedle Investments and Bloomberg, March 2015.

What happens next to labour pricing power is thus extremely important for asset markets. Labour market globalisation has largely been a China story to date. Some IMF work suggests that demographics and the emptying of cheap mobile rural labour may lead to labour shortages in around five years (Figure 11). Supply of non-Chinese cheap labour into the global trading system is dependent upon governance – the delivery of higher education standards, the introduction of market-based economics, and the delivery of trade openness. It is unlikely to be a straight line.

Figure 11. IMF scenarios for China surplus labour (millions)

Source: Das and N’Diaye (2013), Columbia Threadneedle Investments and Bloomberg, March 2015

Even if structurally the rise of China and emergence of the world has fundamentally changed the bargaining power of labour in the west, cyclically, labour’s bargaining power is likely to still be a function of labour scarcity.

It is hard to call a turn in the pricing power of labour, and there has been no real wage response to date. But it is unlikely that we will need to wait until unemployment rates reach zero percent, and on current trajectories, they are on course to hit zero in the US and the UK by 2020 (Figure 12).

Figure 12. US and UK unemployment rates 1980-2015 (and projected on last 12 month trends)

Source: Das and N’Diaye (2013), Columbia Threadneedle Investments and Bloomberg, March 2015

What this means for portfolio construction

Strategic asset allocations for institutional investors have typically included significant exposure to government bonds despite expectations that long-term returns will be below equity returns. Their inclusion in portfolios has been (correctly since 1980, see Figure 13) understood as being an efficient way to reduce aggregate portfolio volatility and increase risk-adjusted returns.

While strong dividend growth might support equity prices, the discount rate tailwinds of the last 35 years will turn to headwinds in an environment of rising (Wicksellian) interest rates associated with an increase in labour power. The last time we witnessed a sustained increase in labour power in the west was between 1945-1980, when cash dominated government bonds on the ex post efficient frontier. Given that the starting point for yields is low, it is conceivable that government bonds will lose their place in institutional portfolios as they stop providing portfolio insurance and returns to incremental interest rate duration turn negative.

Figure 13. Ex-post efficient frontiers between government bonds and equities, and average cash rates, 1945-1980 and 1980-2014

Source: Columbia Threadneedle Investments and Bloomberg, March 2015


Bario, C, P Disyatat and M Juselius (2013)m “Rethinking potential output: embedding information about the financial cycle”, Bank of International Settlements Working Paper, No 404.

Bernanke, B (2015), “Why are interest rates so low?”, Brookings.edu, 30 Mach.

BIS 84th Annual Report (2014), Chapter 4: “Debt and the financial cycle: domestic and global”.

Bonnet O, P Bono, G Chappelle, E Wasmer (2014), “Capital is not back: A comment on Thomas Piketty’s ‘Capital in the 21st Century’”, VoxEU.org, 30 June.  

Broadbent, B (2014), “Monetary policy, asset prices and distribution”, Speech given at the Society of Business Economics Annual Conference.

Das, M and P N’Diaye, (2013), “Chronicle of a Decline Foretold: Has China Reached the Lewis Turning Point?”, IMF Working Paper WP/13/26.

Goodhart, C and P Erfurth, (2014), “Monetary policy, asset prices and distribution”, VoxEU.org, 3 November.

Hisnanick, J and A Rogers, (2007), “Household Income Inequality Measures Based on the ACS Data: 2000-2005”, United States Census Bureau.

Lakner, C and B Milanovic, (2013), “Global Income Distribution, From the Fall of the Berlin Wall to the Great Recession”, World Bank Policy Research Working Paper 6719. 

Lenin, V (2010), Imperialism – the Highest Stage of Capitalism, Penguin Classics.

Milanovic, B (2011), “Global inequality: from class to location, from proletarians to migrants”, World Bank Working Paper 5820.

Milanovic, B (2012), “Global Income Inequality by the Numbers: in History and Now, An Overview”, World Bank Working Paper 6259. 

Piketty, T (2014), Capital in the Twenty-First Century, Harvard University Press.

Teulings C and R Baldwin eds. (2014), Secular Stagnation: Facts, Causes and Cures, VoxEU.org eBook, CEPR Press.


1 See Bario et al. (2013) for an extremely thoughtful analysis.

2 See Chapter 4 BIS 84th Annual Report.

3 Lenin (2010) being the classical example.

4 This debunking of Piketty was first laid out in relation to housing in Bonnet et al. (2014).

5 See, for instance, Broadbent (2014) and Bernanke (2015) on the Wicksellian rate of interest in a contemporary setting.

6 Goodhart and Erfurth (2014).

7 It is interesting to note that three-quarters of the gain in share of GDP from top decile earners in the UK and about half of the gain in the GDP hare from top quintile earners in the US is redistributed down the income spectrum. Pre-distribution real income of the median US household is almost flat, but after the impacts of taxes and benefit changes are felt it is up nicely; participation in US economic growth for the masses has largely been manufactured by the state, somewhat in contradiction to the traditional narrative.

8 In terms of education standards, governance systems, infrastructure, etc.

9,554 Reads