A. J. Brown and the US Phillips Curve: A Comment

Abstract

Several recent papers have focused on Paul Samuelson and Robert Solow’s 1960 article applying the Phillips curve to the US. These interactions have been partly technical in nature, but have also involved discussion of the use of the curve in US policy making and its interpretation. The attention here is on the contribution of Samuelson and Solow’s work in light of prior analysis of the US situation by the English economist A. J. Brown. Much of what Samuelson and Solow argue was already understood by Brown, and his empirical analysis was at least as insightful as theirs.

Keywords:

• Phillips curve
• Paul Samuelson
• Robert Solow
• wage inflation
• unemployment

Acknowledgement

The author would to thank Robert Solow for permission to quote from his email correspondence. Very helpful comments from the editors and referees of the journal resulted in considerable improvements to the paper.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 Brown seems to have missed Tinbergen’s (1951) econometric work on UK business cycles. Tinbergen, though, replaced the employment term with the inverse of the unemployment rate to capture non-linearities.

2 Brown’s ‘critical’ is defined as the unemployment rate below which the ‘unit labour cost of production rises’. This is taken to mean wages rising above productivity increases as with Samuelson and Solow. Brown does not specify his exact assumptions regarding such things as levels of productivity change.

3 To give context to Solow’s comment, it was offered in response to ‘I am also a bit surprised that you explicitly note in your presentation/paper that nothing like Phillips’ work had been done regarding the USA prior to your paper. Brown had actually considered this, again in Chapter 4 of his book. Indeed, he has a figure almost identical to yours’ (Button 2018b).

4 Presumably here, Solow was thinking of either Brown (1938) or (1939).

5 There is a slight contradiction when comparing this to Solow’s quotation cited earlier in terms of who introduced Phillips’ work – Solow or Samuelson?

6 Whether the report itself had a major impact on US policy is doubtful (Schwartz 1987).

7 Both Brown and Samuelson and Solow are lax in their citing of data sources. The latter (p. 187) do allude to ‘Rees’s data’ – presumably Albert Rees, the US labour economist. Some sleuthing by Hoover (2015a, 2015b) and Hall and Hart (2016) suggests sources for the data used in Samuelson and Solow and that it covered 1889–1959.

8 Later, Solow indicated that he and Samuelson were attracted to Phillips’ long-run paper but did not find the same stability in the US series as Phillips had for the UK (Snowdon and Vane 1999). Basically, the same results as Brown.

9 There was some subsequent debate about the exact period that was used in the estimates (Hoover 2015b).

10 Whether it is possible to control for the position of an economy on a particular curve is another matter. The literature on Samuelson/Solow offering a menu of policy options is extensive – Schwarzer (2013) offers a summary and Forder (2014) a critique.

11 As with much of the subsequent literature, the discussants at the AEA meeting seems unsure whether Samuelson and Solow were interested in a menu of policy options or price stability conditions. Lerner (217 in Chandler et al. 1960), for example asks, “How much inflation for how much output or vice versa?” whereas Pechman: (218 in Chandler et al. 1960) talks in terms of “…‘terms of trade’ between unemployment and price stability”

12 Brown does not provide detailed discussion of previous US ‘Price Control’ in his chapter 7.

13 See correspondence between Solow and Hoover (Hoover 2015b, 23).

14 Hoover (2015b) contains calculations and a diagram using US price and unemployment data for 1934–41 and 1946–58; the former being almost exactly the 1935–41 period over which Brown estimated a critical unemployment level for the US of ‘probably over 14%’. Hoover suggests a figure of about 13%. Without the availability of computing facilities in 1955 – probably not even an electronic calculator – Brown’s estimate seems pretty good.

15 This was in response to Button (2020) who had thanked Solow for his earlier comments and sent an initial draft of this paper. ‘I would like to add a note to this debate focusing on Arthur Brown’s work on the US. I attach a draft of my initial thoughts. As I say in it, this reflects a personal interest in the ways economic (and other) ideas are disseminated, why some gain precedence, and in their translation into policy, as well as an interest in Brown’s work.’ Solow’s reply added more detail to earlier correspondence.

Additional information

Notes on contributors

Kenneth Button

Kenneth Button is a University Professor in the Schar School of Policy and Government, George Mason University, Arlington. He has contributed to various aspects of the study of the history of economic thought including The Value of Applied Economics: The Life and Work of Arthur (A.J.) Brown and the editing to the two-volume set, The Economic Works of Harvey Leibenstein.