Abstract
In this paper, we analyze the relationship between industrial stagnation in the US economy and the financialization of nonfinancial firms by asking whether firms in industries experiencing a stronger post-1970 tendency towards stagnation allocate more funds to shareholder payouts and, specifically, share repurchases. While strands of the literature on financialization have long-emphasized the role of stagnation in driving financialization, fewer papers have considered this hypothesis empirically. Our paper speaks to this space in the literature, by linking industrial stagnation in capital accumulation to a firm’s decision to financialize. We, first, use firm-level data to construct an empirical measure of industrial stagnation. Drawing on insights from the Monopoly Capital School, we measure stagnation using the Baran ratio — which describes the average share of surplus allocated towards investment within an industry — and show a secular decline in the average Baran ratio since 1980. Second, we analyze if the tendency towards stagnation captured by the declining Baran ratio predicts the likelihood and magnitude of a firm’s shareholder payouts. We show that firms in industries with a stronger stagnation tendency (a lower Baran ratio) are more likely to repurchase stock and, among firms that do repurchase, that a lower Baran ratio predicts a higher magnitude of these shareholder payouts. These results suggest that a slowdown on the nonfinancial side of the economy is one factor underlying financialized firm behavior in the post-1980 USA.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Xu (2019)’s analysis is at the macroeconomic level and uses the profit share of GDP. We use the profit share of value added to translate this concept to firm-level accounting data.
We drop observations classified as finance, insurance, or real estate by either SIC or NAICS codes (SIC codes 6000–6799, and NAICS codes with first two digits of 52 or 53) and restrict to firms incorporated in the USA and reporting in US dollars (using Compustat variables FIC and CURCD). We, also, drop observations with negative sales, total assets, or financial payouts (repurchases, dividends, or interest payments) (0.07% of the sample), each of which is conceptually restricted to be greater than or equal to zero.
Total debt is the sum of current debt (DLC) and long-term debt (DLTT) if these variables are non-missing; if (at least one of) these variables are missing and total debt (DT) is non-missing, we use DT to define total debt. This definition of the financial profit rate and financial assets (the sum of receivables, RECT; cash and short-term investments, CHE; other current assets, ACO; investments and advances, IVAEQ and IVAO; and other assets, AO) follows Davis (2017).
The choice of deflator does not affect our results.
Following Donangelo et al. (2019), we set \(\Delta INV{G}_{i,t}\) to zero when either \(INVF{G}_{i,t}\) or \(INVF{G}_{i,t-1}\) is missing.
Because the variable XLR includes all labor costs, including those for supervisory workers, it implies an upper (lower) bound on the labor (profit) share.
The employment variable includes all employees of consolidated subsidiaries; part-time and seasonal employees; officers; and full-time equivalent employees. It does not include consultants, contract workers, or employees of unconsolidated subsidiaries. While there is no way to distinguish between part- and full-time employment using Compustat data, Donangelo et al. (2019) show that this measure of average employment provides a good proxy for labor costs. Donangelo et al. (2019) use the Fama–French 17-industry categories; our measure of the Baran ratio is robust to these classifications. We use 2-digit SIC codes to ensure comparability with the rest of our analysis, below.
We, also, compare our main econometric results to an alternative measure of the Baran ratio that does not rely on this imputation method, defined as investment relative to the sum of profits and the change in inventories (see Appendix Tables 8, 10, and 11). In the main body of the paper, however, we define the profit share as the residual of the labor share to explicitly emphasize the role of increased worker exploitation (i.e., stagnant wages despite rising productivity) (Kotz 2008) in driving the stagnation tendency emphasized in this paper.
The first two digits of SIC codes distinguish major groups within economic divisions (e.g., the chemicals industry is a major group within the manufacturing division). We use SIC codes because Compustat has a far higher incidence of missing NAICS codes. However, the regression results are highly robust to 2-digit NAICS classifications.
This trend occurs across sectors; thus, the declining across-industry average is not the artifact of a major decline within a single large industry or sector (such as manufacturing).
In Figure 5 in the appendix, we plot the evolution of the numerator (investment share) and denominator (profit share, or surplus) of the Baran ratio. These trends highlight that, while both a falling numerator and rising denominator contribute to the declining Baran ratio over this period, the profit share plays a particularly key role. Thus, the falling Baran ratio reflects, in large part, a falling labor share of income.
For a survey discussing this institutional evolution in more detail, see Davis (2017).
Net equity issuance is new stock issues (Compustat item SSTK) less gross repurchases (PRSTKC), normalized by total equity (SEQ). To aggregate each component variable, we first sum across all firms in each year, and then construct the final series. This series, therefore, measures net equity issuance at the sector level.
When interpreting magnitudes, note that shareholder payouts are part of a firm’s cash flows and, therefore, are not shares of assets. Normalizing by total assets ensures that the ratios in Fig. 3a are positive for each firm in each year, thereby avoiding complications in interpretation of the results that arise when normalizing by firm-level equity (which can be negative). The overall trends, however, are insensitive to this choice of normalization.
Tax code changes may, also, have played a role in this substitution of share repurchases for dividends. First, capital gains are taxed at more favorable rates than ordinary income (which includes dividend income), creating an incentive for managers to substitute share repurchases for dividends to maximize shareholder returns (Grullon and Michaely 2002). Even after the Tax Reform Act of 1986 reduced the relative tax advantage of capital gains, the gap between these two tax rates remained. Second, a 1993 change to the tax code stipulating a $1 million cap on the tax deductibility of nonperformance-based pay, without a similar cap on “performance-based” pay (including stock-based pay that increases with the stock price) also reinforced growing managerial orientation to shareholder value and the stock price as measures of firm performance (Stout 2012; Davis 2016).
Because a firm can repurchase shares and pay dividends in the same year, the sum of the shares of firms repurchasing shares and paying dividends will generally not equal the share of firms making any shareholder payouts.
We use linear probability models rather than logistic regressions both because they have a more intuitive interpretation and because, in our case, none of the predicted probabilities fall outside of the unit interval, such that the estimates are considered largely unbiased and consistent. Linear probability models with robust standard errors provide a good guide in applications for which the results do not produce a large share of predicted values outside the zero–one range (Cameron and Trivedi 2005; Horrace and Oaxaca 2006). The results are, however, robust to instead using a logistic specification; we report these results in Table 7.
Nonetheless, we avoid making causal claims in the interpretation of these regressions. If, for example, shareholder value orientation is also clustered by industry, then reverse causality problems may persist despite the exclusion of each individual firm from the industry-level measure. In this regard, it is relevant to note that the descriptive evidence in Davis (2016), for example, finds that repurchase growth follows similar trends across industries, suggesting a lack of industry-level clustering. Nonetheless, the extent to which shareholder value orientation is clustered by industry is an important issue for future research aiming to untangle causality in empirical work on financialization.
Time-invariant industry-level factors beyond the Baran ratio, which may also explain a firm’s tendency towards financialization, are absorbed by these firm fixed effects. We include firm-level, rather than industry-level, fixed effects because firm fixed effects also pick up firm-specific factors that are not captured by industry dummies.
The unconditional probabilities in Table 1 differ somewhat from those reported in the descriptive statistics in Appendix Table 6 because the calculations in Table 6 do not restrict to non-missing observations of the regression variables in Columns 1 and 2. These differences in sample do not affect the interpretation of results.
We use a log plus one transformation for debt to avoid dropping observations with zero debt from the regression.
The observation count in Columns 3–4 is higher than in Columns 5–6 due to missing values. To be included in Columns 5–6 either (i) dividends are positive (in which case repurchases can be positive, zero, or missing), or (ii) repurchases are positive (in which case dividends can be positive, zero or missing), or (iii) both dividends and repurchases are zero. This third criterion, which avoids assuming missing values of repurchases or dividends are zero, is stricter than in Columns 1–4 (which require non-missing data on repurchases or dividends). These results are, however, strongly robust to replacing missing values of repurchases and dividends with zeros (i.e., assuming that firms that do not report repurchase or dividend data did not make this type of payment) and, thereby, using identical samples across specifications. We show these results in Appendix Table 9.
Table 7 shows estimates that use a logistic specification; Table 8 measures the Baran ratio as investment relative to the sum of profits and the change in inventories of final goods; and Table 9 shows results that impute zeros for missing values of repurchases and dividends. In the interest of space, we only include results for the contemporaneous version of the dependent variable.
To construct this variable, we assign any missing observations a value of zero to avoid restricting the sample to firms with non-missing values of repurchases for all 3 years (\(t\), \(t+1\), and \(t+2\)).
In Tables 10 and 11, we show these results are robust to instead measuring the Baran ratio as the ratio of investment to the sum of profits and the change in inventories of final goods for both the contemporaneous dependent variable (Table 10) and the 3-year moving average (Table 11). While the coefficients describing the contemporaneous link between this alternative measure of the Baran ratio and repurchases are statistically insignificant in Table 10, the sign of the estimate is preserved and total payouts are significant. The 3-year estimates in Table 11, also, retain their statistical significance, reiterating that a dynamic readjustment towards repurchases takes place after a decline in the Baran ratio. Finally, note that, as we restrict the sample used for the fixed effects regressions to non-zero observations of shareholder payouts, we do not include results imputing zeros for missing values of shareholder payouts.
References
Aglietta M (2000) Shareholder value and corporate governance: some tricky questions. Econ Soc 29(1):146–159
Aglietta M, Bretton R (2001) Financial systems, corporate control and capital accumulation. Econ Soc 30(4):433–466
Akkemik K, Özen S (2014) Macroeconomic and institutional determinants of financialization of non-financial firms: case study of Turkey. Soc Econ Rev 12(1):71–98
Baran P, Sweezy P (1966) Monopoly Capital. Monthly Review Press, New York
Barradas R, Lagoa S (2017) Financialization and Portuguese real investment: A supportive or disruptive relationship? J Post Keynesian Econ 40(3):413–439
Belotti F, Deb P, Manning W, Norton E (2015) twopm: two-part models. The Stata Journal 15(1):3–20
Bhargava A (2010) An econometric analysis of dividends and share repurchases by US firms. J R Stat Soc Ser A Stat Soc 173(3):631–656
Boyer R (2000) Is a finance-led growth regime a viable alternative to Fordism? A Preliminary Analysis. Econ Soc 29(1):111–145
Cameron C, Trivedi P (2005) Microeconometrics: Method and applications. Cambridge University Press, New York
Crotty J (2003) The neoliberal paradox: the impact of destructive product market competition and impatient finance on nonfinancial corporations in the neoliberal era. Rev Radic Polit Econ 35(3):271–279
Davis L (2016) Identifying the ‘financialization’ of the nonfinancial corporation in the US economy: a decomposition of firm-level balance sheets. J Post Keynesian Econ 39(1):115–141
Davis L (2017) Financialization and investment: a survey of the empirical literature. J Econ Surv 31(5):1332–1358
Davis L (2018) Financialization and the non-financial corporation: an investigation of firm-level investment behavior in the United States. Metroeconomica 69(1):270–307
Davis L, Orhangazi Ö (2020) Competition and monopoly in the U.S. economy: what do the industrial concentration data show? Compet Chang 25(1):3–33
Deb P, Norton EC (2018) Modeling health care expenditures and use. Annu Rev Public Health 39(1):489–505
Demir F (2009) Financial liberalization, private investment, and portfolio choice: financialization of real sectors in emerging markets. J Dev Econ 88(2):314–324
Despain H (2015) Sweezyian financial instability hypothesis—Monopoly Capital, inflation, financialization, inequality and endless stagnation. Int Crit Thought 5(1):67–79
Donangelo A, Gourio F, Kehrig M, Palacios M (2019) The cross-section of labor leverage and equity returns. J Financ Econ 132(2):497–518
Dünhaupt P (2017) Determinants of labour’s income share in the era of financialization. Camb J Econ 41(1):283–306
Fligstein N (1990) The Transformation of Corporate Control. Harvard University Press, Cambridge
Foster J (2008) The financialization of capital and the crisis. Mon Rev 59(11):1–19
Greenwald D, Lettau M, Ludvigson S (2019) How the wealth was won: factor shares as market fundamentals. NBER Working Paper No. 25769
Grullon G, Michaely R (2002) Dividends, share repurchases, and the substitution hypothesis. J of Fin 57(4):1649–1684
Grullon G, Larkin Y, Michaely R (2019) Are U.S. industries becoming more concentrated? Rev Finan 23(4):697–743
Guschanski A, Onaran Ö (2018) The labor share and financialisation: evidence from publicly listed firms. Greenwich Papers in Political Economy No GPERC59
Horrace W, Oaxaca R (2006) Results on the bias and inconsistency of ordinary least squares for the linear probability model. Econ Lett 90(3):321–327
Jiang Z, Kim K, Lie E, Yang S (2013) Share repurchases, catering, and dividend substitution. J Corp Finan 21:36–50
Kohler K, Guschanski A, Stockhammer E (2019) The impact of financialisation on the wage share: a theoretical clarification and empirical test. Camb J Econ 43(4):937–974
Kotz DM (2008) Contradictions of economic growth in the neoliberal era: accumulation and crisis in the contemporary U.S. economy. Rev Radic Polit Econ 40(2):174–188
Lapavitsas C (2011) Theorizing financialization. Work Employ Soc 25(4):611–626
Lazonick W (2009) Sustainable prosperity in the new economy? Business organization and high-tech employment in the United States. Kalamazoo: W.E. UpJohn Institute
Lazonick W (2014) Innovative enterprise and shareholder value, Academic-Industry Research Network (AIR) Working Paper #14–03/01
Lazonick W, O’Sullivan M (2000) Maximizing shareholder value: a new ideology for corporate governance. Econ Soc 29(1):13–35
Lin KH, Tomaskovic-Devey D (2013) Financialization and U.S. income inequality, 1970–2008. Am J Sociol 118(5):1284–1329
Magdoff F, Foster J (2014) Stagnation and financialization: the nature of the contradiction. Monthly Review 66(1):1–24
Magdoff H, Sweezy PM (1987) Stagnation and the financial explosion. Monthly Review Press
McDonough T, Reich M, Kotz K (2010) Contemporary capitalism and its crises: social structure of accumulation theory for the 21st century. Cambridge University Press
Onaran Ö, Stockhammer E, Grafl L (2011) Financialisation, income distribution and aggregate demand in the USA. Camb J Econ 35(4):637–661
Orhangazi O (2008) Financialisation and capital accumulation in the non-financial corporate sector. Camb J Econ 32(6):863–886
Seo H, Kim H, Kim J (2016) Does shareholder value orientation or financial market liberalization slow down Korean real investment? Rev Radic Polit Econ 48(4):633–660
Skinner DJ (2008) The evolving relation between earnings, dividends, and stock repurchases. J Financ Econ 87(3):582–609
Stephens C, Weisbach M (1998) Actual share reacquisitions in open-market repurchase programs. J Financ 53(1):313–333
Stockhammer E (2004) Financialisation and the slowdown of accumulation. Camb J Econ 28(3):371–404
Stockhammer E (2005) Shareholder value orientation and the investment-profit puzzle. J Post Keynesian Econ 28(2):193–215
Stout L (2012) The shareholder value myth: how putting shareholders first harms investors, corporations and the public. Berett-Koehler Publishers Inc, San Francisco
Tori D, Onaran Ö (2018) The effects of financialization on investment: evidence from firm-level data for the UK. Camb J Econ 42(5):1393–1416
Tori D, Onaran Ö (2020) Financialization, financial development and investment: evidence from European non-financial corporations. Soc Econ Rev 18(3):681–718
Van Treeck T (2008) Reconsidering the investment-profit nexus in finance-led economies: an ARDL-based approach. Metroeconomica 59(3):371–404
Xu Z (2019) Economic surplus, the Baran ratio, and capital accumulation. Mon Rev 70(10):25–39
Author information
Authors and Affiliations
Corresponding author
Additional information
We would like to thank Randy Albelda for detailed comments on previous drafts of this article.
Appendices
Appendix A
The investment share of value added and the profit share (1971–2019). a Investment share of value added. b The profit share. Notes: This figure disaggregates the numerator of the Baran ratio (capital expenditure relative to value added) in panel (a) from the denominator (the profit share of value added, defined as one minus the labor share of income) in panel (b). Each ratio is trimmed at the 1st and 99th percentile to account for outliers. For variable definitions, see Sect. 2
Appendix B, robustness checks
Rights and permissions
About this article
Cite this article
Davis, L., McCormack, S. Industrial stagnation and the financialization of nonfinancial corporations. Rev Evol Polit Econ 2, 459–491 (2021). https://doi.org/10.1007/s43253-021-00043-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s43253-021-00043-6