Understanding Weak Capital Investment: the Role of Market Concentration and Intangibles Nicolas Crouzet and Janice Eberly

Size: px

Start display at page:

Download "Understanding Weak Capital Investment: the Role of Market Concentration and Intangibles Nicolas Crouzet and Janice Eberly"

Stewart Campbell
5 years ago
Views:

1 Understanding Weak Capital Investment: the Role of Market Concentration and Intangibles Nicolas Crouzet and Janice Eberly Questions Findings 1. Why is capital investment low? - 5percentagepointsbelowpre-2 strend - Taking into account corporate valuations (Tobin s Q), cash flow, cost of capital, etc. 2. Why have US industries become more concentrated? - Sales Herfindahl index has increased by at least 5% in 75% of US industries since mid-199 s - Superstar firms, or market power? 1. Physical investment is low because of the rising importance of intangible capital - R&D, software, business processes - An omitted factor implies a wedge - Treating intangibles as an omitted factor can account 3 to 6% of the investment shortfall 2. Intangible capital has economic properties that may have enabled the rise in concentration - Non-rivalry: economies of scale - Excludability: market power - Intangible intensity correlated with market share - Heterogeneous e ects across industries, notably: /Retail: scale e ects (productivity) : market power (markups) 1/1

2 Since the early 2 s, investment has been low relative to corporate valuations and other fundamentals. 5 I/K (%) There has been a large and increasing shortfall of the investment rate relative to fundamentals since 21. This investment gap is visible in aggregate (top panel), but and also in four leading sectors (bottom panel). The gap ranges from 5 percentage points (Manufacturing) to 11 percentage points () /- 2 s.e. I/K (%) 5 Manufacturing /1

3 This investment gap is highly correlated with the share of intangibles in total capital Intangible capital has become a larger share of firms capital stocks over time (top panel). This is especially in true some industries, such as and (see the top panel of page 6). Fraction of total assets Balance sheet intangibles (Compustat, aggregate; LHS) Capitalized software and IP (BEA, RHS) Across industry/years, the investment gap is larger, the larger the share of intangibles. (Bottom panel; each point is an industry/year.) 5 I/K (%) Balance sheet intangibles as a share of total assets (average, Compustat) Investment gap (p.p.) Fitted values 3/1

4 Q-theory indicates that there should be a wedge between investment and Tobin s Q if a capital input is omitted. - In a standard Q-theory model, investment in physical capital, K follows: i t = 1 1 g 1+r E t qk,t+1 P K,t, q K,t = V t (marginal q) (1) K t - If the model is augmented to include intangible capital B t as an input in production, standard linear homogeneity considerations imply that: Tobin s Q = Q t = V t K t = q K,t + B t K t q B,t. (2) - Therefore, Q will overstate the true incentive to invest in physical capital. - The investment-q relationship is: i t = 1 1 g 1+r E t [Q t+1 ] P K,t {z } standard investment predictors B t 1 K t 1+r E t qb,t+1 {z } investment gap (3) - There should be an investment gap, and the gap should be larger, themorethe firm or the industry relies on this other factor precisely the negative correlation observed between the investment gap and the intangible share in the data. 4/1

5 Accounting for the wedge created by intangibles, the investment gap is substantially smaller. 5 - The residual investment gap (green line) is substantially smaller than the raw investment gap (blue line) after controlling for industry-level intangible shares. - The changing composition of the capital stock of US industries can account for about 6% of the total investment gap (top panel). The results are consistent across industries (bottom panel). I/K (%) /- 2 s.e. - Direct firm-level (unweighted) estimates of the investment-q relationship, controlling for the intangible share, suggest a somewhat lower number, of about 3%. - Also, the investment gap is largest among the most intangible-intensive firms, even within industries. I/K (%) Manufacturing /1

6 Intangibles are associated with larger firms and more concentrated industries. - Afirm smarketshareinitsindustryishigher when its intangible share is higher. - This relationship holds between firms of the same industry, within firms over time, and controlling for year e ects. Dependent variable : market share (A) (B) (C) Compustat intangible share 38 ***.96 ***.73 *** (17.69) (5.4) (4.91) Observations Industry year f.e. Yes No No Firm f.e. No Yes Yes Year f.e. No No Yes - Both within detailed industries, and aggregating across them, the intangible share of capital has risen as industries have become more concentrated over time goods Manufacturing Herfindahl index of sales (LHS) Balance sheet intangibles (Compustat, RHS) 6/1

7 Why should concentration be related to intangibles? Intangible capital has unusual economic properties, which may have enabled the increase in concentration. - Di erent types of intangible capital could be captured by the rising intangible share: Patents, software, business processes - A common feature of these di erent types of capital is that they are non-rival Scale economies =) high measured labor productivity - Another feature of (some) of these types of capital is that they are excludable, oratleastnoteasilyappropriatedorreplicatedbycompetitors Di erentiation and/or to barriers to entry =) high markups and/or profit margins 7/1

8 In and, to a lesser extent,, intangible investment is empirically associated with higher markups Estimated markups have risen considerably in and. They were stable in Manufacturing and /Retail Manufacturing - These increases are strongly correlated with the increase in intangibles, especially in and also in. - Because of potential measurement issues, we instrument the balance sheet intangible share to improve the estimation. The results also hold using operating profits to measure markups Markups from firm-level data Markups adjusted to match average Hall (218) industry estimates Dependent variable : markup (log) Cross-sectional (between) regressions - At the industry level, the results are large: intangibles can account for up to two thirds of the increase in markups. Intangible share s j,t (IV) 231 *** *** (6.81) (5.92) First-stage F stat Panel (within) regressions Intangible share s j,t (IV).723 *** 4.59 *** (3.41) (2.8) First-stage F stat /1

9 In /Retail and, to a lesser extent, in, intangible investment is associated with higher labor productivity. - Since 199, estimated labor productivity has increased twofold in the /Retail sector and fivefold in. - The increase in productivity in these industries is strongly correlated with intangible investment Manufacturing The results are robust to instrumenting the balance sheet intangible measure, and hold both across and within firms. - At the industry level, the results are also large: intangibles can account for up to three quarters of the increase in labor productivity in the sector. Sales per worker (Compustat) Output per hour (BEA) Dependent variable : productivity (log) Cross-sectional (between) regressions Intangible share s j,t (IV) *** (.85) (2.37) First-stage F stat Panel (within) regressions Intangible share s j,t (IV) ***.834 *** (2.79) (2.72) First-stage F stat /1

10 Conclusions Findings 1. The growing importance of intangible capital intellectual property, software, business processes canaccountfortheongoingweaknessin physical investment, both theoretically and empirically. 2. Intangible capital has economic properties that may have enabled the rise in concentration - Non-rivalry: economies of scale - Excludability: market power 3. Intangible intensity is higher, and growing more quickly, among market leaders. In certain sectors, such as, it is primarily associated with growing market power. In others, like /Retail, it is primarily associated with higher productivity. Policy issues 1. Investment is usually associated with embodied technological change, so slowdown is worrisome, but less so if rising intangibles embody new tech. 2. Intangibles may be less sensitive to usual policy levers. 3. Developing common valuation frameworks for intangibles would support market development and improve collateral use. 4. But if new technologies are embodied in intangibles, understanding market power versus productivity is key. Open questions 1. What are the boundaries of intangible capital? Is it defined by its function brand, logistics or by its economic characteristics returns to scale, excludability? 2. Who appropriates the returns to intangible capital? Does intangible capital reinforce a declining labor share? 1 / 1

INTANGIBLE CAPITAL: IMPLICATIONS FOR INVESTMENT AND MARKET STRUCTURE. Janice Eberly 1,2

INTANGIBLE CAPITAL: IMPLICATIONS FOR INVESTMENT AND MARKET STRUCTURE Janice Eberly 1,2 1 Kellogg School of Management, Northwestern University and NBER 2 Based on research with Nicolas Crouzet, Kellogg