Presidential Address: Pension Policy and the Financial System

• Author: DAVID S. SCHARFSTEIN
• DOI: http://doi.org/10.1111/jofi.12710
• Date: 01 August 2018
• Published date: 01 August 2018

The Journal of Finance R

David S. Scharfstein

The President of the American Finance Association 2017

THE JOURNAL OF FINANCE •VOL. LXXIII, NO. 4 •AUGUST 2018

Presidential Address: Pension Policy

and the Financial System

DAVID S. SCHARFSTEIN∗

ABSTRACT

In this paper, I examine the effect of pension policy on the structure of financial

systems around the world. In particular, I explore the hypothesis that policies that

promote pension savings also promote the development of capital markets. I present a

model that endogenizes the extent to which savings are intermediated through banks

or capital markets, and derive implications for corporate finance, household finance,

banking, and the size of the financial sector. I then present a number of facts that are

broadly consistent with the theory and examine a variety of alternative explanations

of my findings.

COUNTRIES AROUND THE WORLD have made vastly different choices about how to

meet the retirement needs of their populations. Some have chosen to finance

retirement incomes largely by taxing current workers, in so-called pay-as-you-

go (“PAYGO”) pension systems. Others have promoted private pension saving as

a way to fund retirement benefits. For example, according to the OECD, in Italy

and Denmark the average worker entering the labor force in 2014 could retire

at age 67 and expect to have almost 70% of his or her preretirement income

replaced by pension income (OECD (2015)). In Italy this pension income would

be financed in large part by taxes on current workers, while in Denmark much

of this income would be generated by the assets of private pension schemes.

These different approaches are reflected in differences in pension assets across

the two countries: In 2014, private pension assets were less than 10% of gross

domestic product (GDP) in Italy, while in Denmark they were about 180% of

GDP (OECD (2015)). Table I, described in greater detail in the next section,

documents the remarkable variation in pension systems across the OECD.

∗David S. Scharfstein is at Harvard Business School and NBER. Presidential Address delivered

to the American Finance Association, Philadelphia, January 6, 2018. I am very grateful to Will

Diamond, Robin Greenwood, Sam Hanson, Gianluca Rinaldi, Jeremy Stein, and Adi Sunderam for

many valuable discussions and for helpful comments on drafts of the paper. I also benefited from

discussions with Hugo B¨

anziger, John Beshears, John Campbell, Garrett Curran, Sarah Feldman,

Andrea Hamaui, Victoria Ivashina, Anil Kashyap, Divya Kirti, David Moss, Thomas Philippon, Jim

Poterba, TarunRamadorai, AndreiShleifer, Paul Tucker,Boris Vall´

ee, Luis Viceira, the passengers

of Carpe Diem V,and seminar participants at Yale Law School and the Chicago Booth Initiative on

Global Markets. I thank Andi Wang for providing extraordinary research assistance throughout

the project. I am also grateful to James Palano and Francesca Guiso for their help at various stages

of the research. The Division of Research at Harvard Business School provided generous funding

for this work. Disclosure: I am a director of M&T Bank Corporation, which offers banking, asset

management, and retirement products and services.

DOI: 10.1111/jofi.12710

1463

1464 The Journal of Finance R

Tabl e I

Summary Statistics of Pension Variables

This table presents characteristics of pension systems in 23 OECD countries in 2014. All data are reported in percentages. The public (private)

pension replacement rate is the OECD’s forecast of the percentage of lifetime average preretirement income provided by the public (private) pension

system for a retired worker with mean preretirement income. A private pension replacement rate of zero indicates that the OECD does not calculate

a replacement rate because coverage is not broad, although there may be some private pensions in the country. Public (private) pension assets are

assets held by the public (private) pension system. The mandatory pension contribution rate is the OECD’s estimate of the percentage of an average

worker’s wage that is required to be contributed by an employer or an employee to a private or public pension scheme. See the Appendix for detailed

variable definitions and sources.

Pension Replacement Rate (%) Pension Assets/GDP (%) Mandatory Pension Contribution Rate (%)

Country Public Private Total Public Private Total

Public,

Employee

Public,

Employer

Private,

Employee

Private,

Employer

Australia 13.5 31.0 44.5 5.9 115.2 121.1 0.0 9.5 0.0 0.0

Austria 78.1 0.0 78.1 0.0 6.4 6.4 10.3 12.6 0.0 0.0

Belgium 46.6 13.3 59.9 5.2 5.2 10.4 7.5 8.9 0.0 0.0

Canada 36.7 29.3 66.0 17.1 137.3 154.5 5.0 5.0 0.0 0.0

Denmark 21.5 46.3 67.8 0.0 186.2 186.2 0.5 0.8 0.0 12.0

Finland 55.8 0.0 55.8 27.6 54.6 82.2 7.1 17.8 0.0 0.0

France 55.4 0.0 55.4 2.5 7.9 10.5 6.8 8.5 3.0 3.0

Germany 37.5 12.5 50.0 0.0 6.1 6.1 9.5 9.5 0.0 0.0

Greece 66.7 0.0 66.7 0.0 0.6 0.6 6.7 13.3 0.0 0.0

Iceland 3.4 65.8 69.2 0.0 141.1 141.1 0.0 7.8 4.0 8.0

Ireland 34.7 30.4 65.1 0.0 53.3 53.3 4.0 10.8 0.0 0.0

Italy 69.5 0.0 69.5 0.0 7.6 7.6 9.2 23.8 0.0 0.0

Japan 35.1 0.0 35.1 25.2 17.9 43.2 8.7 8.7 0.0 0.0

Netherlands 27.1 63.4 90.5 0.0 145.7 145.7 4.9 0.0 16.0

New Zealand 40.1 12.4 52.5 10.3 19.8 30.1 0.0 0.0 3.0 3.0

Norway 44.0 5.8 49.8 5.8 7.5 13.3 8.2 14.1 0.0 0.0

Portugal 73.8 0.0 73.8 7.0 9.8 16.8 6.4 13.8 0.0 0.0

Spain 82.1 0.0 82.1 5.4 13.3 18.7 4.7 23.6 0.0 0.0

Sweden 37.0 19.0 56.0 28.7 66.9 95.6 7.0 11.4 0.0 4.5

Switzerland 23.3 16.9 40.2 0.0 111.8 111.8 4.2 4.2 7.7 10.4

Turkey 75.7 0.0 75.7 0.0 4.4 4.4 9.0 11.0 0.0 0.0

UK 21.6 29.8 51.4 0.0 92.9 92.9 9.1 11.9 0.0 0.0

USA 35.2 32.6 70.3 15.9 136.8 152.7 6.2 6.2 0.0 0.0

Mean 44.1 17.8 62.0 6.8 58.6 65.4 5.9 10.1 1.5 1.9