How realistic are return expectations, really?

How realistic are return expectations, really?

08-02-2018 | 月次アウトルック

Equities and real estate can structurally deliver returns in excess of economic growth, but within limits, says Robeco investor Lukas Daalder.

  • Lukas Daalder
    Chief Investment Officer

Speed read

  • US pension funds assume extremely high future returns
  • Past returns boosted by high dividends, rising earnings
  • Neither seem to be repeatable in the decades to come

Two recent research papers triggered the question of whether it is possible for equities and real estate to structurally produce returns higher than underlying economic growth. Although this is indeed feasible, future returns may not be as generous as they have been in the past, says Daalder, Chief Investment Officer of Robeco Investment Solutions.

The first research paper, entitled ‘The Return Expectations of Institutional Investors’, looked at the long-term expectations of 230 US pension funds1. According to the study, their nominal average expected return is 7.6%, resulting in an expected real return of 4.8%. This is based on long-run average nominal returns on cash (3.2%), bonds (4.9%), real estate (7.7%), hedge funds (6.9%), publicly traded stocks (8.7%) and private equity funds (10.3%).


Somewhat ambitious

“This looks pretty ambitious given that historically, OECD data shows that the average annual realized return that these US pension providers over 2006-2016 was 1.5% in nominal terms and -0.3% in real terms,” Daalder says. “And that was despite the average total return of the S&P 500 coming in at 6.9% during this timeframe.”

“Given the continued decline in productivity seen in recent decades, as well as the popularity of the secular stagnation school of thought, it raises the question of how realistic the return expectations for the riskier parts of their portfolio really are,” says Daalder.

“By way of comparison, Robeco’s own long-term (steady state) expected nominal return on equities is 7%, while bonds are expected to yield 4.25%. As such, you still need a pretty optimistic view of the world to get to a 7.6% longer-term return envisaged by these pension funds. Even when ignoring the early February stock market correction, recent performance clearly does not support these kinds of return expectations.”

Link with economic growth

Daalder says it begs the question of whether there is a direct link between economic growth and asset returns that would make forecasting more reliable. He says the answer to that question partly comes from the second piece of research, entitled ‘The rate of return on everything, 1870-2015’. In this study, the authors look at the historical track record of the nominal and real returns of 16 developed nations since 18702.

“The picture in this study is clear: with the exception of the two decades that marked the world wars, returns have been quite clearly in excess of the underlying real growth rates of the 16 countries studied,” Daalder says. “The average real growth rate over the period was 3.1%, while the diversified portfolio return was 5.9%. But there is a catch.”

“Looking at the breakdown of the portfolio used, it is clear that all of this so-called ‘excess’ return came from the risky assets: the returns on cash (1.3%) and bonds (2.5%) on average lagged the growth of the real economy, while the returns on stocks (7.0%) and real estate (6.7%) exceeded it.”

“Based on these historic returns, the 4.8% assumed real return of pension funds all of a sudden does not appear to be too outlandish after all. However, this outcome raises a number of questions, led by asking how is it possible that returns on risky assets can structurally outstrip growth.”

Role of dividends

Part of the answer lies in the role that dividends have played in generating the ‘excess’ return made in equities and real estate in the past, Daalder says. According to the Shiller database, the total geometrical nominal return for the US has been 8.9% since 1871, while the average annual dividend yield over that timeframe has been 4.4%.

“As such, this seems to be a valid and stable reason to expect returns in excess of growth, which can be seen as compensation for the risks involved with equities compared to risk-free assets such as bonds and cash,” he says.

“The story doesn’t end there, however. Stock prices have also risen more than the underlying growth rate of the economy, adding to the excess return recorded in the past. This is partly because there is a structural mismatch between the earnings growth of listed companies – which only represent a small subset of the economy – and wider economic growth.”

“Meanwhile, higher levels of leverage, and exposure to growth outside the 16 reported countries (the emerging markets), are all factors that can lead to higher earnings growth compared to GDP growth.”

Too much wealth will lead to lower future returns

Daalder says another problem with future expected returns is that rising PE levels over many decades mean that stocks have structurally become more expensive over time.

“You only need to look at the very low yields that we currently see in the bond market to make the point: these low future returns have come on the back of above-average returns as bond prices were bid up. You see the same process in real estate and equities: you get above-average returns on your investment as stocks and houses are being bid up, but this leads to a reduction in future dividend yield or rental returns.”

“Which brings us back to dividends being an important part of the excess return we have seen in the past. The long-term dividend yield has been 4.4% for the US, but if we look at the current dividend yield of the S&P 500, this has declined below the 2% level. Given the importance of dividends in the ‘excess’ return, it is pretty safe to say that from this starting level, one should not expect the same returns that have been reached in the past.”

1Andonov, Aleksandar and Rauh, Joshua D., The Return Expectations of Institutional Investors (September 29, 2017). Stanford University Graduate School of Business Research Paper No. 18-5. Available at SSRN: https://ssrn.com/abstract=3091976 or http://dx.doi.org/10.2139/ssrn.3091976

2Jordà, Òscar and Knoll, Katharina and Kuvshinov, Dmitry and Schularick, Moritz and Taylor, Alan M., The Rate of Return on Everything, 1870–2015 (December 2017). NBER Working Paper No. w24112. Available at SSRN: https://ssrn.com/abstract=3089509


当資料は情報提供を目的として、Robeco Institutional Asset Management B.V.が作成した英文資料、もしくはその英文資料をロベコ・ジャパン株式会社が翻訳したものです。資料中の個別の金融商品の売買の勧誘や推奨等を目的とするものではありません。記載された情報は十分信頼できるものであると考えておりますが、その正確性、完全性を保証するものではありません。意見や見通しはあくまで作成日における弊社の判断に基づくものであり、今後予告なしに変更されることがあります。運用状況、市場動向、意見等は、過去の一時点あるいは過去の一定期間についてのものであり、過去の実績は将来の運用成果を保証または示唆するものではありません。また、記載された投資方針・戦略等は全ての投資家の皆様に適合するとは限りません。当資料は法律、税務、会計面での助言の提供を意図するものではありません。




商号等: ロベコ・ジャパン株式会社  金融商品取引業者 関東財務局長(金商)第2780号

加入協会: 一般社団法人 日本投資顧問業協会