Macro Roundup

Nicolas Crouzet and Janice Eberly, "Rents And Intangible Capital: A Q+ Framework," National Bureau Of Economic Research, July 2021, https://www.nber.org/papers/w28988

Comparison to existing literature“…These findings are qualitatively consistent with the recent literature arguing that pure profits as fraction of value added have been growing over the last three decades (Guti´errez and Philippon, 2017; Barkai, 2020; Karabarbounis and Neiman, 2019). However, they differ quantitatively. For instance, Barkai (2020) finds that the pure profit share rose from -5.6% in 1984 to 7.9% in 2014, an increase of 13.5 p.p. over the period. Karabarbounis and Neiman (2019), in their “case⇧,” find that the pure profit share must have risen by about 13 p.p. over the same period. We find an increase in rents of half that magnitude…”

On the paper’s limits, “…However, it has three limitations. First, it does not allow for non-convex adjustment costs. Second, it abstracts from financial constraints. The next subsection discusses extensions in this direction. Third, it assumes that rents, µ, are exogenous. In particular, they do not depend on past investment, in contrast, for instance, with models of customer capital.9 In this sense, our results are restricted to “neoclassical” models of the firm, and provide a benchmark against which the effects of other frictions on the investment gap can be compared….”

“… From the standpoint of the model, these changes are driven by three underlying forces,reported in Figure 2: a greater importance of intangibles in the production function; higher rents; and a decline in user costs, more pronounced for physical than for intangible capital.The top left panel of Figure 2 shows that even using the relatively narrow definition of intangibles in the NFCB data, the share of intangible capital in production,⌘, increased substantiallyafter 1985, from 0.17 to 0.29 in 2015.25 The behavior of the intangible share approximately mimics the behavior of the measured ratio of intangible to physical capital at replacement cost, which increases rapidly after 1985. The effects of the intangible share on the overall investment gap are magnified by the rise in rents after 1985. The top right panel of Figure 2 reports estimates of the rents implicit in Equation (13). In order to facilitate comparison with existing estimates, we express them as the flow value of rents relative to value added, which is related to the parameter controlling rents in the model...Rents, as a fraction of value added, increase from 1.5% in 1985, to 7.7% in 2015 — a cumulative 6.2 percentage point (p.p.) change over three decades. Expressed as markups over value added, this is an increase from 1.015 in 1985, to 1.083 in 2015. Finally, we note two other features of our time-series for the investment gap. First, the gap is elevated during the 1960s; the decomposition attributes this to a combination of low user costs (driven by the low interest rates of the period), and high rents.27 Second, the gap is particularly small during the 1975-1985 period. The model primarily attributes this reversal to the large increase in discount rate and the decrease in growth rates around the early 1980s, which, by reducing the present value of future rents, pushes the average value of installed capital closer to its marginal value….”

Specifically"...from a theoretical perspective, we show that the gap between average Q and marginal q for physical capital, which we call the “investment gap”, can be decomposed into three distinct terms: a term capturing rents to physical capital, a term capturing the value of installed intangible capital, and a term capturing rents to intangible capital. The last element of this decomposition, an interaction term that is new to our analysis, is particularly important: it clarifies the fact that rising rents and rising intangibles cannot be meaningfully analyzed in isolation, as their interaction contributes to the gap between investment and returns. Moreover, this decomposition is very general, as our framework nests a number of existing investment model..."

Second:"..we show that this interaction term is empirically important to the recent rise in the investment gap. Importantly, we show how each term in our decomposition can be quantified using data on profits, investment, valuations, and estimates of the intangible capital stock within the structure of the model. In aggregate data, the interaction term accounts for between one-quarter and one-half of the investment gap, depending on how broad the definition of intangibles is. In addition, our approach leads to lower estimates of the increase in total rents than existingwork. As we show, this is equivalent to a smaller estimate of the decrease in total user costs of capital. This occurs because while including intangibles raises valuations, it also boosts the user cost of capital due to higher depreciation rates (hence reducing rents)...."

Evidence, “…Figure 1 reports the investment gap decomposition... The decomposition emphasizes three main finding. First, the investment gap is large during two distinct periods: 1960-1970, and after 1985. The wedge between average Q and marginal q is therefore not strictly a hallmark of the post1980s period. Second, rents attributable to physical capital — the first term in Equation (13) — play a sizable (though somewhat declining) role in explaining the investment gap: they account 61% of it in 2015, compared to 67% in 1965.24 Third, rents attributable to intangibles — the third term in Equation (13) — have become markedly more important in recent years. In 2015, 25% of the investment gap reflects the combined effects of high rents and a large stock of intangibles, compared to 10% in 1965, using the BEA measure of R&D capital only, the narrow measure of intangibles available in these data….”

Core takeaway, "...This research provides a general decomposition of the gap between average Q — which is observable — and marginal q — the shadow value that drives investment. This decomposition captures the effects of unmeasured capital, such as intangibles, and also the effect of rents. We use measurement of the gap to shed light on the growing divergence between physical investment and valuations, which our approach interprets as being driven by the combined effects of growing rents and growing intangible capital. With a relatively narrow measure of intangibles (R&D capital), one-third of the investment gap reflects a combination of growth in the intangible capital stock and rents generated by intangible capital. Expanding the definition of intangibles beyond R&D increases this contribution to about two thirds. In addition to these aggregate effects, sectoral results show that rents on intangibles are largest in some of the fastest growing sectors in the economy, such as Tech and Health, and that within these sectors, rents are highest in subsectors with rapid growth in intangibles, as well…”

New NBER argues that the capital investment shortfall in the US has been driven by the rise of intangible capital which allow firms with market power to extract more rents. They estimate that 1/3 of the gap in capital investment is driven by intangibles particularly in fast-growing sectors, "... the divergence between returns and investment can be cast as a rising gap between the average value of business capital, or Tobin’s average Q, and its marginal value, or Tobin’s marginal q. We directly observe rising average Q in the data, via market values, while marginal q is a shadow value measured implicitly by lackluster investment. A gap between the average value of capital and its marginal value can arise and grow for a number of reasons..."

Research from the Boston Fed ~ directionally mirroring Natasha Sarin’s work finds that including Social Security and retirement wealth in the wealth concept results in significantly lower top shares and less growth in the concentration of wealth,

Their core takeaway, “…Focusing on two pre-retirement age groups from 1989 to 2019,we find that using an expanded measure of wealth that includes DB pension and Social Security resources to estimate wealth concentration in the United States results in a level that is lower than estimates based on more commonly used measures of market net worth. We present this finding through several wealth concentration measures, including shares of wealth held by the top of the distribution, ratios of wealth held at different points in the distribution, and the Gini coefficient. We also find that while wealth concentration has risen over the past three decades, it has done so more slowly for this expanded wealth measure….”

The data, “….To present an alternative and, we think, useful addition to the measurement of wealth concentration, we use the Survey of Consumer Finances (SCF) to develop an expanded measure of wealth that incorporates both estimates of defined benefit (DB) wealth and the expected present value of Social Security among the 40- to 59-year-old population. We directly incorporate the work of Sabelhaus and Volz (2019), who impute the value of DB wealth of current workers in the SCF using labor market and pension plan characteristics in the survey along with high-quality external data on DB plan assets. In this section, we discuss the SCF and our methods for (1) estimating earnings histories of survey respondents, (2) calculating future Social Security benefits, and (3) estimating age-forward SCF net worth, as well as DB pensions, to the point of retirement….”

Findings, “…Our estimates show that the value of DB pensions and Social Security are significant relative to other forms of wealth—throughout the wealth distribution but especially at the lower half of the wealth distribution. Indeed, we find that, even for the median household, the present value of DB pensions and Social Security benefits accounts for more than half of all wealth. With respect to their effects on the distribution of wealth, we find that (1)including DB pension and Social Security wealth results in markedly lower measures of wealth concentration, and (2) trends toward higher wealth inequality over time, while moderated, are still present. In particular, the “90/50 ratio”—the ratio of wealth held by those at the 90th percentile of wealth to those at the 50th percentile—is reduced by nearly half for the 50- 59 age group (from 13.4 to 6.8 in 2019) and for the 40-49 age group (10.7 to 6.4) when we include the estimated value of Social Security. The “50/10 ratio” declines even more with the inclusion of Social Security; for 2019, the ratio falls from 13.1 to 4.3 among those aged 40 to 49 and from 21.3 to 4.2 for the 50-59 age group. The share of wealth held by the “top 5 percent” drops from about 72 percent down to 51 percent when defined contribution (DC) plan and DB pension wealth are added to non-retirement wealth; it falls even further, to 45 percent, when Social Security benefits for those aged 40 to 59 are included. The inclusion of each measure, however, has a somewhat different effect: Social Security decreases wealth concentration “at the top,” whether we look at the top 5 percent’s share of wealth or the 90/50 ratio; DB decreases the top 5 percent’s share, but in more recent years, it actually increases the 90/50 ratio. The top 5 percent’s share of our expanded wealth measure rises 8 fewer percentage points compared with the top 5 percent’s share of non-retirement wealth over the 1989-2019 period….”

Impact on Wealth Inequality, “….We find that wealth inequality rose over the 1989-2019 period and that the inclusion of Social Security and retirement plan wealth has an impact on both the level of inequality and its trend. In this section, we focus on ratios of wealth held at the 90th percentile of the wealth distribution to wealth held at the 50th percentile (P90/P50), as well as ratios of wealth held at the 50th percentile to wealth held at the 10th percentile (P50/P10), as measures of the skewness of the wealth distribution. In Figure 3, we see that among the 40-49 age group, the P90/P50 of non-retirement wealth rose from 4.9 in 1992 to 6.0 in 2019; among 50- to 59-year-olds, it climbed from 5.2 to 8.0. The P90/P50 of expanded wealth for the younger age group rose from 3.9 in 1992 to 6.4 in 2019. For 50- to 59-year-olds, the combined wealth P90/P50 rose from 3.4 in 1989 to 6.4 in 2019. In contrast, the P50/P10 ratio of expanded wealth declined slightly for 40- to 49- year-olds, from 4.4 to 4.3 from 1992 to 2019. It also decreased for 50- to 59-year-olds, falling from 4.9 to 4.2 over the same time period….. In Table 2, we show the top 5 percent’s and top 10 percent’s shares of different wealth concepts for another perspective on wealth concentration. Here, we see that a large share of all types of wealth are held at the top of the distribution, but especially so for non-retirement wealth. For instance, those in the top 10 percent of the non-retirement wealth distribution among 40- to 49-year-olds held 60 percent of non retirement wealth in 1989 and 75 percent in 2019. Including Social Security and retirement wealth in the wealth concept results in significantly lower top shares and less growth in the concentration of wealth. For all households in our sample, we estimate that in 2019 the top 5 percent of the distribution held 72 percent of non-retirement wealth but 51 percent of wealth including DB and DC pensions and only 45 percent of expanded wealth, which includes net Social Security wealth.12 Social Security is very broadly held and has an equalizing effect; the top 5 percent held only 8 percent of this component in 2019. From 1989 to 2019, the top 5 percent’s share of non-retirement wealth rose 18 percentage points, while its share of expanded wealth rose only 10 percentage points. Similar trends are seen when comparing the share of wealth held at the top 5 percent and top 10 percent of the distribution for both age groups over time: Concentration decreases substantially when DB and Social Security wealth are considered, with increases in wealth concentration remaining over time for all wealth concepts…”

Lindsay Jacobs, Elizabeth Llanes, Kevin Moore, Jeffrey P. Thompson, and Alice Henriques Volz, “Wealth Concentration in the United States Using an Expanded Measure of Net Worth,” Federal Reserve Bank Of Boston, June 2021, https://www.bostonfed.org/publications/research-department-working-paper/2021/wealth-concentration-in-the-united-states-using-an-expanded-measure-of-net-worth.aspx

McKinsey data on intangibles, core findings, "...Even through economic disruptions, intangibles investments have increased. Over the past 25 years, the United States and ten European economies (Austria, Denmark, Finland, France, Germany, Italy, the Netherlands, Spain, Sweden, and the United Kingdom) achieved 63 percent growth in gross value added (GVA), a measure of economic growth. During this period, the investment share of intangibles increased by 29 percent. Rising investment in intangibles has been associated with increasing total factor productivity of entire economies. Growth in investment in intangibles slowed after the global financial crisis, and productivity growth decelerated, too, suggesting a link. Investing in intangibles correlates with productivity and sector growth. In the past quarter century, intangibles investment has risen in all sectors, and data from INTAN-Invest indicates that there is an observable link between investment in intangibles and GVA growth. It also indicates a strong association with total factor productivity. Sectors that have invested the most in intangibles— more than 12 percent of their GVA— achieved 28 percent higher growth than other sectors in GVA, at more than 2.7 percent per year between 1995 and 2019. The relationship is strongest in knowledge-intensive services such as financial services and in innovation-driven services such as telecommunications, media, and technology. This apparent correlation reflects the synergistic nature of different types of intangibles. Companies with top-notch digital analytics attract the best talent, and that talent improves the quality and scope of the analytics. Reflecting such synergies, companies and sectors that invest across intangible categories post higher GVA than their peers. Regardless of the sector, companies that invest more in intangibles grow more. The new survey indicates that top growers, defined as companies in the top quartile of GVA growth by sector in 2018-19 (whose median growth was 20 percent) are investing 2.6 times more in intangibles than low growers, defined as the bottom 50 percent of companies for GVA growth in 2018-19 (whose median growth was 3 percent).The gap increases to between five and seven times in sectors such as financial services where competitive advantage is anchored in knowledge..."

“…Most other sectors invested less than 12 percent of their GVA in intangible capital and have achieved below-average rates of GVA growth, at 2.1 percent per year. In general, a relationship can be observed between higher intangibles investment and higher growth, although there are some exceptions. One exception is manufacturing, which has invested 15.3 percent of its GVA in intangibles but has experienced broadly flat growth. This appears to reflect the fact that manufacturing has invested particularly heavily in innovation capital, specifically R&D, which entails large sunk costs as companies seek to develop the right new products. Another example is healthcare (Exhibit 6)….”

“…Three economies stand out for having achieved both high investment in intangibles and robust growth in GVA: Sweden, the United Kingdom, and the United States. Sweden achieved annual 2.4 percent GVA between 1995 and 2020, the United States 2.3 percent, and the United Kingdom 2.2 percent. All three economies have larger shares of knowledge and innovation sectors, in which intangibles investment is prominent. Conversely, economies that posted lower rates of growth in GVA over this period were more exposed to resource- or labor-intensive sectors that tend to invest less in intangibles. Many other factors determine growth, of course, and they should be examined in detail, but there does appear to be a link between investment in intangibles and GVA growth at the whole-economy level. An intangibles-rich economic model is not the only way for an economy to promote productivity and growth. Nevertheless, economies that are experiencing growth in intangibles investment are also posting growth in total factor productivity (Exhibit 4…”

Edward Glaeser, and James Poterba take the lay of the land on America's infrastructure, their bottom line, "...Our assessment of the role of economic analysis in infrastructure investment suggests several broad conclusions. First, it is difficult to place high confidence in widely discussed measures of infrastructure “need.” The most reliable way to develop such estimates would be by applying cost-benefit analysis on a project-by-project basis and aggregating the results. But that approach is expensive, given the vast array of potential infrastructure projects, and it is subject to gaming by overstating future benefits and low-balling costs. Estimates of the returns to maintaining existing infrastructure are often higher than estimates of the returns to undertaking new projects, which suggests the importance of guarding against “ribbon-cutting bias” toward new initiatives on the part of both elected leaders and the heads of government agencies. Any major infrastructure initiative should emphasize careful ex ante analysis of project costs and benefits, with oversight where feasible of padding by advocates of the assumptions regarding future costs and benefits. Second, infrastructure projects in the United States are expensive relative to those in other nations. The precise reasons are difficult to identify, but they include project designs that incorporate many features that remediate adverse project effects, such as highway noise and the inconvenience of disruption while building, required wages for workers that may exceed area norms, project delay through regulatory processes, and weak procurement and project management by the relevant government agencies.Third, user fees warrant greater consideration as a source of infrastructure project financing. Such fees, along with congestion charges, can improve the efficiency of infrastructure use. While there are concerns about the distributional effects of user fees and burdens on low-income groups in particular, the pattern of infrastructure use across income groups suggest that some user fees are progressive—higher income households use airports, for example, more than their lower income counterparts. Public transit, particularly buses, is a notable exception. Rather than carry out income redistribution by exempting infrastructure use from charges, policymakers could consider targeted redistribution programs, such as transit vouchers for low-income households or infrastructure-use rebates mediated through the tax system. Some states currently provide income tax relief for renters or for commuters who can document their travel costs. Finally, public-private partnerships can provide a means to increase operational efficiency, but arguments that they allow project sponsors to access low-cost capital should be viewed with caution. In some cases, the cost of capital for private entities may exceed that for public sector borrowers and relying on private finance rather than public funding may ultimately increase the 42 cost of the project. Some state and local governments may be attracted to these partnerships because they relieve current cash flow constraints, but they may come at a price in terms of the long-term cost of infrastructure services...."

Edward Glaeser, and James Poterba, "Economic Perspectives on Infrastructure Investment," Aspen Economic Strategy Group, July 14, 2021, https://www.economicstrategygroup.org/publication/poterba-glaeser-infrastructure-investment/

Evidence:“…Flyvbjerg, Bruzelius, and van Wee (2008) compare the capital costs for urban rail projects around the world. The costs for the six systems in the United States that were included in the analysis range from $88 million per kilometer (Atlanta) to $147.5 million per kilometer (Baltimore). Thirteen out of 17 of the European systems in the study, and five out of six in Asia or Latin America, had costs below $88 million per kilometer. Levy (2011) argues that these comparisons understate the cost disadvantage of U.S. projects, noting that “the American projects examined are quite old, from the 1980s, and many have large above-ground parts.” He further identifies three New York City projects with costs of $1.3 billion, $1.7 billion, and $4 billion per km, as well as San Francisco’s Central Subway, which cost $500 million per kilometer even though, as a light rail tunnel, it was a less demanding project. Levy’s updated Transit Cost database reports actual or projected cost-per-kilometer data, converted to $US using purchasing power parity exchange rates, on 540 different projects, including 256 that were completed by 2020. We inflation-adjust these cost estimates using the CPI and assume that the median dollar was spent in the year that was half-way between the start and end date of the project. We treat projects with average years beyond 2021 as having an average year of 2021. 7 Table 4 presents our findings.For the 19 projects in the database that are in the United States, the average cost was $1,601 million per mile, compared with a non-U.S. global average of $478 million. The median U.S. project was $965 million per mile, compared with a non-U.S. median of $299 million. The database also contains information on the share of the rail system that is underground. When we restrict our analysis to the 255 projects that are 100% in tunnels, the median cost of the 11 U.S. projects is $1,379 million per mile, compared with a non-U.S. global median of $341 million. While precise comparisons are difficult, many projects in densely populated foreign cities have substantially lower per-mile costs than their U.S. counterpartsThe cost breakdown highlights a number of important patterns. First, real estate costs are a tiny share of the project’s total cost, despite New York City’s sky-high property prices. Second, neither administrative and regulatory costs, nor the even smaller category of environmental mitigation (contained within construction and only $2.14 million) were significant causes of the high costs. Third, the two largest elements in construction costs were tunneling ($3.1 billion) and stations and intermodal facilities ($2.3 billion). The very expensive station construction is one reason why East Side Access was the most expensive project in the database,but the tunneling on its own is extraordinarily costly by global standards. While direct environmental mitigation itself was a small component of the accounting costs, environmental factors play a much larger role in the overall cost of the project by changing the nature of construction itself. For example, BVP (2018) explain that “the Environmental Impact Statement (EIS) required construction activities in Manhattan to take place in the subterranean realm, with almost all equipment and spoils transitioning through the 63rd Street tunnels to staging sites in Sunnyside Yards,” which typically meant “laborers filling burlap bags with spoils that were then loaded onto trains to Queens (or in some cases, the Bronx) and then unloaded and sorted by laborers.” The Metropolitan Transit Authority (MTA) estimated that it could have saved $75 million in schedule-related costs alone by deploying a simpler system similar to that used in other projects…Arguably, the conditions for tunneling in Manhattan are as difficult as anywhere in the world, although cost estimates for projects in London, which are all completely underground, are only one-third as high as those in New York City. Labor costs are higher in the United States than elsewhere, and especially so in New York City, but this reflects institutions as well as generally high labor costs. The Bureau of Labor Statistics (BLS) reports that in May 2020, the median hourly wage for electricians in the New York City metropolitan area was $36.13, and the mean was $40.48.9 BVP (2018) report a minimum hourly wage for electricians of $65 on the East Side Access project and an added $62 dollars per hour in benefits, making the per-unit labor cost for the project a multiple of the prevailing wage….”

Why does it cost so much to build infrastructure in the U.S.?“…Beyond accounting, there are three deeper explanations for why infrastructure costs are so high in the United States. First, it is possible that conditions are more demanding and that raw materials and labor are more expensive than elsewhere. This explanation suggests that higher costs are unavoidable but should be considered in discussions of optimal infrastructure policy, since higher costs are a good reason to build less infrastructure. Second, it is possible that the agencies charged with building infrastructure are poorly designed to manage costs. In this case, there may be changes to infrastructure building practices that could lower costs and stretch infrastructure budgets. Third, it is possible that external factors, especially the threat of litigation or political backlash, lead to expensive forms of mitigation, which change the nature and cost of building projects(Altshuler and Luberoff 2005)…”