Tag: investment

Investment, Animal Spirits and Algae

Arjun and I did a webinar recently on our book Against Money, organized by Merijn Knibbe. We’re very grateful to him for putting it together, and should have video to share soon.

Even in a friendly setting like this, it can be a challenge to explain what the real-world stakes are in debates over money. But as it happens, there was a Matt Levine column the same day as the webinar, that offers a perfect application of one of the central themes of the book.

To be honest, this is not really surprising. You could even think of our project as backfilling the economic theory behind Levine’s columns, which the textbooks certainly don’t help with. “How Keynes explains last week’s Money Stuff” could be an elevator pitch for the book.

The lead item in this Money Stuff was about a hypothetical algae farming startup, and the financing thereof:

You start a startup with a far-fetched idea like genetically engineering algae to produce clean renewable fuel. You go out to investors to raise money. You say “we are going to genetically engineer algae to produce clean renewable fuel, if we succeed we will make a bajillion dollars, you want in?” The investors think that sounds cool, because it does. But they are responsible investors, they do their due diligence, they ask questions like “is that a thing” and “can you actually produce fuel algae” and “will it be cost-effective?” You do your best to answer their questions.

Do you exaggerate? Oh sure. That is the job of a startup founder. I once wrote, approximately:

What you want, when you invest in a startup, is a founder who combines (1) an insanely ambitious vision with (2) a clear-eyed plan to make it come true and (3) the ability to make people believe in the vision now. “We’ll tinker with [algae] for a while and maybe in a decade or so a fuel-[producing strain of algae] will come out of it”: True, yes, but a bad pitch. The pitch is, like, you put your arm around the shoulder of an investor, you gesture sweepingly into the distance, you close your eyes, she closes her eyes, and you say in mellifluous tones: “Can’t you see the [algae producing clean fuel oil] right now? Aren’t they beautiful? So clean and efficient, look at how nicely they [float in this pond], look at all those [genes], all built in-house, aren’t they amazing? Here, hold out your hand, you can touch the [algae] right now. Let’s go for a [swim].”

Of course, you are a startup founder; you are in essence a salesperson. Back at the lab, the algae scientists and chemical engineers and accountants are looking at your pitchbook in disbelief. “Wait, you’re telling investors that we can produce the fuel oil now? You’re telling them that we’ll have large profits in two years? Did you not read our latest status report?” The scientists and accountants are boring and conservative; it is their job to try to make the dream work in dreary reality. It is your job to sell the dream now.

(The brackets are there because he is repurposing text from an earlier column on AI.)

This is a story about finance, not venture capital specifically. The details would be different if the algae company were getting a loan from a bank, but the fundamental situation would be the same.

I want to make a few points about this.

First, what’s being described here is not a market outcome. Nobody has yet purchased any fuel made from genetically modified algae. To the extent there are market signals here, they point in the wrong direction — at current prices, the cost of producing this fuel would be greater than what it would sell for. Nor has this business shown profits in the past — it’s a startup. Right now, the market is saying this is a value-subtracting activity. Funding it anyway is the opposite of what market signals are saying to do.

Funding the algae project is an explicit decision by someone in authority. It is a decision based on promises. It is based, precisely as Levine says, on dreams.1

Joseph Schumpeter compared the function of banks under modern capitalism to Gosplan, the central planning agency of the old Soviet Union. Banks, through a conscious, deliberate decision, dedicate some fraction of society’s resources to some project that they have decided is worthwhile. “The issue to the entrepreneurs of new means of payments created ad hoc” by the banks, he writes, is “what corresponds in capitalist society to the order issued by the central bureau in the socialist state.”

What’s more, as Arjun and I write in Against Money, banks

are stronger in a certain way than any real central planner, because they have the authority to redistribute anything. A Soviet planner might assign a plant this many tons of some raw material, that much electricity, use of those parts of the transportation network. Money as the universal equivalent is a token granting the holder use of whatever they need. A loan then is a ticket to the entrepreneur saying, you have the authority to take whatever labor and other resources your project requires.

In this sense, markets are not an alternative to planning, they are a tool for planning. Money is the substrate within which planning takes place.

People used to talk about a “soft budget constraint” as a defining feature of the Soviet economy — enterprises could continue operating even if their costs exceeded their sales, as long as the planners saw some social value in their continued operation.2 Startups like the algae power company have the softest of budget constraints — they are able to incur substantial costs, often over many years, without any sales at all.

This is not some weird quirk of venture capital. This is a central purpose of finance – to direct society’s resources to one activity that has not yet been successful in the market, but that somebody think could be. The defining characteristic of an entrepreneur is that they undertake some new activity, something that is not already being done, with funding provided by someone else. An entrepreneur in this sense definitionally faces a soft budget constraint.

This is not, again, an anomaly, it is not a breakdown of the normal operation of capitalism. It is essential to what makes capital such a powerful force for transforming our material existence. And it needs to be central to our theoretical accounts of capital and of the investment process.

It certainly was for Keynes. As he famously observed in Chapter 12 of the General Theory,

a large proportion of our positive activities depend on spontaneous optimism rather than on a mathematical expectation, whether moral or hedonistic or economic. Most, probably, of our decisions to do something positive, the full consequences of which will be drawn out over many days to come, can only be taken as a result of animal spirits—of a spontaneous urge to action rather than inaction, and not as the outcome of a weighted average of quantitative benefits multiplied by quantitative probabilities.

Enterprise only pretends to itself to be mainly actuated by the statements in its own prospectus, however candid and sincere. Only a little more than an expedition to the South Pole, is it based on an exact calculation of benefits to come. Thus if the animal spirits are dimmed and the spontaneous optimism falters, leaving us to depend on nothing but a mathematical expectation, enterprise will fade and die;—though fears of loss may have a basis no more reasonable than hopes of profit had before.

It is safe to say that enterprise which depends on hopes stretching into the future benefits the community as a whole. But individual initiative will only be adequate when reasonable calculation is supplemented and supported by animal spirits.

Markets and the pursuit of private profit have existed for much longer than the their fusion with long-lived means of production command over wage labor that we call capital. One important reason for the failure of profit-seeking, through most of its history, to revolutionize production, is that these activities were subject to hard budget constraints and forced to adhere closely to market signals. Through most of their history, they couldn’t create new forms of production on the basis of dreams.

The algae company is getting access to real resources — authority over other people’s labor — because they have convinced a planner that their project is worthwhile.

Market socialists — whose belief in the virtue of markets is exceeded only, perhaps, by 19 year olds who have recently discovered Ayn Rand — like to ask how socialism can maintain the material accomplishments of capitalism without markets. But it isn’t markets that that produce the genuine and immense material accomplishments of capitalism.

The initial investments in AI or algae farming — or automobiles or airplanes or antibiotics — are not a response to market signals. They are conscious choices by some group of people to try something that hasn’t been done before. We might like algae and dislike AI (I do), but the solution is some substantive improvement in the planning system. It’s not an issue of planning versus markets.

Now, some people might say: This planning is based on the hope of future profit, it will eventually have to be validated by markets. But it is not incidental that the market outcome and the pursuit of profit are mediated by conscious planning.. They do not happen automatically. The judgement of the market can be deferred, in principle indefinitely.

We must also reject the idea that the assessment of future profitability is rational or objective. This is one reason the Levine story is useful – it focuses our attention on the ways that financing decisions are made in practice. Making energy from algae is cool! As he says, this an important part of the investment process. That should not be abstracted from.

There are many potentially profitable businesses that never get access to financing. The required return for most startups is very high, or effectively infinite. Manias may be essential to maintain an adequate level of investment. The irrationally high discount rate applied to future returns can only be offset by an irrationally high expectation of future profits. (See, as for much of this post, the current AI boom.)

Nor is it clear that future profit always is the motivation, certainly not the only one, and certainly in the early stages. It’s not incidental that Levine emphasis that algae energy could get funding in part because it is cool. It’s not, perhaps, incidental that OpenAI started its existence as nonprofit. The pursuit of profit is not always what motivates investment, especially when it involves fundamental departures from existing forms of production.

This conflict between the pursuit of profit and large-scale fixed investment goes back to the beginning of industrial capitalism. As Eric Hobsbawm observes in his classic account of the Industrial Revolution, the textile industry — small scale, labor-intensive — could develop through largely self-financed improvements on existing production methods serving existing markets.But the large-scale capital-goods industry, using novel techniques to serve a market that was only brought into existence by the Industrial Revolution itself, was a different story. There, the pursuit of profit was an inadequate spur in the absence of some additional non-pecuniary motive.

No industrial economy can develop beyond a certain point until it possesses adequate capital-goods capacity. … But it is also evident that under conditions of private enterprise the extremely costly capital investment necessary for much of this development is not likely to be undertaken… For [consumer goods] a mass market already exists, at least potentially: even very primitive men wear shirts or use household equipment and foodstuffs. The problem is merely how to put a sufficiently vast market sufficiently quickly within the purview of businessmen.

But no such market exists, e.g., for heavy iron equipment such as girders. It only comes into existence in the course of an industrial revolution (and not always then), and those who lock up their money in the very heavy investments required even by quite modest iron-works … are more likely to be speculators, adventurers and dreamers than sound businessmen. In fact in France a sect of such speculative technological adventurers, the Saint-Simonians, acted as chief propagandists of the kind of industrialization which needed heavy and long-range investment.

Th Saint-Simonians driving the investment boom of the 19th century, the rationalists and long-termists and Zizians driving investment in the 21st — perhaps it’s not such a far-fetched analogy. (Though personally I find Saint Simon more appealing.) However different the content, they are filling the same essential function. And that is the key point here — a system that relies on private initiative for irreversible commitments to projects that transform production, cannot be based on rational calculation, on objective market signals. The market outcomes of these kinds of projects cannot be known until long after the die is cast. A different kind of motivation is needed.

A related point: Nobody knows, right now, if the algae thing will work. Nobody knows if AI will turn out to be useful (I think not, or not very, but I am well aware I could be wrong.) The tradeoff is not about allocating real resources to their best use, among the known uses available. If the algae thing doesn’t get funding — and we can be sure that many, many projects as well founded are not getting funded — the reason will not be because society had a more urgent use for those resources. It will be because people couldn’t figure out a way to cooperate — that the mechanisms to convert promises (or dreams) into command over labor did not operate in that case.

(A flip side of this vision, which I can’t go into here but is essential to the larger argument, is that society has resources to spare. Many people’s time is being spent much less usefully than it could be.)

There’s another, more subtle point. It is not just that we don’t know how profitable these projects will be until someone finances them and they are carried out. There is not any fact of the matter about how profitable these projects will be, independent of how they are financed.

This is the point where Arjun’s and my argument may be challenging for a certain strand of Marxists. (It is not, I think, a challenge to Marx himself, who said a lot of different things on these questions, at different levels of abstraction.)

There is an idea — Anwar Shaikh offers a contemporary example — that the rate of profit is determined first, and then the rate of interest is secondary, a special case of profit, governed by it, or a deduction from it. But we can’t say what the profitability of the algae business even is, prior to the question of what terms it is financed. At one rate of interest it may be very profitable, at another less so or not worthier pursuing at all.

Now maybe you will say: sure, anyone can make a profit if they get that free Fed money. But it’s not just that. The relative profitability of different projects depends on the term on which they can be financed.

Let’s consider two projects. One will make energy from burning oil, the other from growing algae. The oil project is straightforward: 100 dollars laid today will yield 120 dollars worth of fossil-fuel energy a year from now. The algae project requires a lot more upfront costs — you have to first, you know, figure out how to make energy from algae. But your best guess is that $100 invested today will allow you to produce $50 worth of fuel from $10 worth of inputs every year starting 15 years from now.

So, which of these two projects should you commit your capital to? Which of them is more profitable?

The answer, of course, is that you can’t say until you know what terms the projects will be financed on.

Partly this is just a simple matter of discount rates. In these narrow terms, the algae project is more profitable if the interest rate is 5 percent; the fossil-fuel project is more profitable if the interest rate is 10 percent.

More broadly we have to consider, for instance, whether the financing will have to be rolled over, if, say, the project takes longer than expected. What are financing conditions are likely to be at that point? If the loan is due and can’t be rolled over and the project has not generated sufficient returns to repay it, then the return on whatever capital the undertaker put in themselves will be negative 100 percent. The chance of this happening — which, again, depends as much on future financial conditions as on the income generated by the project itself — has to be factored in to the expected returns.

We also have to consider the terms of the financing — what kind of collateral will be required? Will it have to be periodically marked to market? What control rights are demanded by investors or lenders? The viability of the project from the point of view of the person carrying it out depends as much on these considerations as on the physical problem of converting algae to energy.

I recall a Wall Street Journal article years ago – I’m sorry, I don’t have a link – on the economics of putting power plants on barges. There are technical issues pro and con, but the decisive advantage of putting a plant on a barge is that it is better collateral. Lenders are more willing to finance a power plant when they can physically tow it away in the event of default.

So if we are going to evaluate the profitability of a power plant on a barge versus one on land, we have to consider how important it is to keep lenders happy — how scarce or abundant financing is. We also have to consider other monetary factors. A big utility, or one guaranteed by a state, can be counted on to pay its debt, so collateral is less important than it is for a smaller business without public backing.

Another way of looking at this is that the distribution of profits has a variance as well as a mean. How much the higher moments matter, depends how confident we are that contracts will be honored in alls states of the world. It depends on how confident we are that short-term deficits can be financed and that only the long-term outcome matters.To the extent that that’s true, we should just focus on mean expected profits. But if defaults are possible, then the higher moments matter too — again complicating the question of what it means for one project to be more profitable than another.

This is the fundamental point Hyman Minsky was making with his two-price model. It’s why he insisted that money is not neutral. The price of long-lived assets depends on the interest rate (or as he put it, the supply of money), in a way that the price of current output does not. The price of a factory relative to the stuff coming out of it will shift as money becomes scarcer or more abundant.

And of course it’s not just two prices. It’s a whole set of prices, for capital goods that are more and less long-lived and are more or less specialized to particular production processes. The more scarce money is, the higher will be the price of the power plant on the barge relative to the power plant on land.

Again, this is not just a time discount. It’s a discount for uncertainty. It’s a discount for commitment. It’s a discount on hopes and dreams versus money on the table.

For every interest rate there is a different schedule of labor values. For every interest rate there is a different set of market signals. A tight-money market socialism does different things from a loose-money market socialism.

This is a version of Sraffa’s argument that one can’t calculate labor inputs for different commodities unless we already know the profit rate, which must be determined from outside the production process, for instance “by the rate of money interest.” Even if we assume that all production possibilities are already known and available, we can’t decide which are most profitable unless we know the terms on which production will be financed.

In the real world, again, the possibilities for production are not known in advance. And contrary to Sraffa’s preferred assumption of content returns to scale, industrial production tends to have increasing returns, implying the existence of multiple equilibria. But directionally, all these considerations point the same way. Easy money makes projects with longer-term returns, higher-variance or more uncertain returns, more specialized capital goods, more increasing returns, and greater departures from current production processes more attractive. Tight money, the opposite.

A central function of discourse around finance, and the stock market in particular, is to obscure this role of finance in shaping and directing production. The stock market creates the situation it pretends to reflect, in which one production process can be smoothly traded off against another.

If the algae-company investment is successful, it will eventually result in the creation of a listing on a stock market, creating a tradable claim on the future profits from algae trading. At that point, income from algae energy will have a market price reflecting its exchangeability with all sorts of other incomes. You will be able to swap one future dollar of algae-energy income with a future dollar of income from any of thousands of other listed companies. It is tempting to treat this as simply a fact of nature, to retroactively project it back to the whole process of building this company, and treat it all as a process of market exchange just like swapping one share for another.

That the delimitation of exchangeability is a distinct problem from the allocation of real resources — that, in a sense, is what our book is about.

Democratizing Finance

(This is the text of a talk I gave for a workshop organized by the International Network for Democratic Economic Planning. The video of the conference is here.)

The starting point for this conversation, it seems to me, is that planning is everywhere in the economy we already live in.

There’s a widespread idea that production today is largely or entirely coordinated by markets. This idea  is ubiquitous in economics textbooks, of course; it also forms a major part of unspoken economic common sense, even for many socialists and others on the left politically. But it seems to me that when you look at things more critically, the role of market coordination in the economies that we live in is in fact rather limited.

Within the enterprise, markets are almost nonexistent. Production is organized through various forms of hierarchy and command, as well as through intrinsic motivation — what David Graeber calls everyday communism or what we might call the professional conscience — the desire to do one’s job well for its own sake.

The formation, growth and extinction of enterprises, meanwhile, is organized through finance. People sometimes talk about firms growing and dying through some kind of Darwinian process, but the function of finance is precisely to prevent that. By redistributing surplus between firms, finance breaks the link between the profits a firm earned yesterday and the funds available for it to invest today.

The whole elaborate structure of banks, stock markets, venture capital and so on exists precisely to make funds available for new firms, or firms that have not yet been profitable. We see this very clearly in Silicon Valley, as in the current boom in “AI” investment — this is as far as you can get from a world where growth is the result of past profits.

On the other side, institutions like private equity, and the market for corporate control, ensure that that the surplus generated in one firm need  not be reinvested there. It can be extracted — consensually or otherwise — and used somewhere else.

In both cases, this is not happening through any kind of automatic market logic, but through someone’s conscious choice.

Once we think of finance as a system of planning , it is natural to ask if it can be redirected to meet social needs, such as addressing climate change. I want to make four suggestions about how we can pursue this idea most effectively.

First. We need to think about where financing constraints matter, and where they don’t.

Many firms do fund investment largely from their own profits; in others, investment spending is modest relative to current costs. In both these cases — where investment is internally financed, and where investment requirements are low relative to costs of production — finance will have limited effects on real activity.

Where finance is most powerful is in new or rapidly growing, capital-intensive sectors, especially where firms are relatively small. Green energy is an important example — for wind or solar power, almost all the costs are upfront. Housing is also an area where finance is clearly important – while this is of course, a very old sector, firms are relatively small, capital costs are large, assets are very long-lived, and there is a significant lag between outlays and income. It is clear that booms and busts in housing construction have a great deal to do with credit conditions.

Labor intensive sectors like care work, on the other hand, are poor targets for credit policy, since costs and revenues occur more or less simultaneously, and capital needs are minimal. Subsidies or other “real” interventions are needed here.

Large, established firms are also likely to be fairly insensitive to credit policy. There’s a great deal of evidence that the internal discount rates corporations use to evaluate investment projects are not tightly linked to interest rates. At best, financing may relax an external constraint where decision makers already operate with long horizons. But what we know about corporate investment decisions suggests that they are not much affected by credit conditions — something that thoughtful central bankers have long understood.

Second. Channeling credit to constrained areas will have a bigger impact than penalizing credit to unwanted areas.

This seems like an important limitation on the types of green policies adopted by the ECB, for example. For firms that issue bonds, the interest rate they face is not likely to be a major factor in their investment decisions. Where credit matters most is for smaller, bank-dependent firms and households, which face hard limits on how much they can borrow.

This is even more the case for the stock market. Firms for which stock issuance is a significant form of financing make up a very, very small group. In general, changes in stock ownership will have no effect on real investment at all.

Related to this is the question of rules vs discretion. It is relatively easy to write rules for what not to invest in. Targeting finance-constrained sectors requires more strategic choices. So this is an instrument that is state-capacity intensive. In a setting of limited capacity, credit policy is unlikely to work well.

Similarly, if we want to see across-the-board changes, as opposed to fostering new growth in particular areas,  credit is not the right tool. In that case it is better to directly regulate the outcomes we are interested in. If you want higher wages, write a minimum wage law. Don’t tell your central bank to penalize holdings of shares in low-wage firms.

Third. We need to think carefully about what parts of finance we want to socialize, and where new institutions are needed and where they aren’t.

Various financial institutions offer funding to real activity (directly or indirectly) on their asset side, while issuing liabilities that some particular group of wealth owners wants to hold. In the case of many institutions — banks, insurance companies, pension funds — their social value comes as much or more from the distinctive liabilities they issue, as from the activities that they finance.

It’s natural to imagine public finance in similar terms, and think of a public investment authority, say, issuing distinctive liabilities that are somehow connected to the activities that it finances. I think we need to tread very cautiously here. The connections between the two sides of private balance sheets are largely irrelevant for the public sector.

The public sector already finances itself on the most favorable terms of any entity in the economy. The private sector’s need for retirement security and other forms of insurance can be addressed by the public sector directly. Public provision of new assets for retirement saving would be a step backward from current systems of public provision.

There is a case for a larger public role in the payments system, and in the direct provision of banking services to those who currently lack access to them. But there is no reason to link this service provision to public credit provision, and a number of good reasons not to.

The stronger arguments for socializing finance, it seems to me, lie on the asset side of the public-sector balance sheet. We don’t need to find new ways of financing things the public already does. We do need to bring public criteria into the financing of private activity.

It’s worth emphasizing that what matters is what gets financed, and on what terms. Who owns the assets has no importance in itself. Setting up a sovereign wealth fund does nothing to socialize investment, if the fund is operated on the same principles as a private fund would be.

I observed this first-hand some years ago, when I worked in the AFL-CIO’s Office of Investment. The idea was to use the substantial assets of union-affiliated pension funds to support labor in conflicts with employers. But in practice, the funds were so constrained both by legal restrictions and by the culture of professional asset management that it was effectively impossible to depart from the conventional framework of maximizing shareholder value.

Fourth. We need to link proposals for socializing finance to a critique of conventional monetary policy. We need to challenge the sharp lines between planning, prudential regulation, and monetary policy proper. In reality, every action taken by the central bank channels credit towards some activities, and away from others.

One important lesson of the past 15 years is the limits of conventional monetary policy as a tool for stabilizing aggregate demand. But central banks do have immense power over the prices of various financial assets, and monetary policy actions have outsized effects on credit-sensitive sectors of the economy. A program of using credit policy for what it can do — fostering the growth of particular new sectors and activities — goes hand in hand with not using credit policy for what it cannot do — stabilizing inflation and employment. In this sense, socializing finance and developing alternative tools for demand management are complementary programs. Or perhaps, they are the same program.

It’s worth noting that Keynes was very skeptical of the sort of fiscal policy that has come to be associated with his name. He did not believe in running large fiscal deficits, or boosting demand via payments to individuals. For him, stabilizing demand meant stabilizing investment spending. And this meant, above all, reorienting it way from future profitability, which is inherently unknowable, and beliefs about which are therefore ungrounded.

This is a key element in the Keynesian vision that is often overlooked: Our inability to know the future matters less when we are focused on providing concrete social goods. It may be very hard, even impossible, to know how much the apartments in a given building will rent for in thirty years, depending as it does on factors like the desirability of the neighborhood, how much housing is built elsewhere, and the overall state of the economy. But how long the building will stand up for, and how many people it can comfortably house, are questions we can answer with reasonable confidence.

Wouldn’t it be simpler, then, to stabilize private demand in the first place, rather than try to offset its fluctuations with changes in the interest rate or public budget position? From this point of view, our current apparatus of monetary policy would be rendered unnecessary by a program of reorienting investment to meet real human needs.

UPDATE: I have added a link to the video of the conference.

Varieties of Industrial Policy

I was on a virtual panel last week on industrial policy as derisking, in response to an important new paper by Daniela Gabor. For me, the conversation helped clarify why people who have broadly similar politics and analysis can have very different feelings about the Inflation Reduction Act and similar measures elsewhere. 

There are substantive disagreements, to be sure. But I think the more fundamental issue is that while we, inevitably, discuss the relationship between the state, the organization of production and private businesses in terms of alternative ideal types, the actual policy alternatives are often somewhere in the fuzzy middle ground. When we deal with a case that resembles one of our ideal types in some ways, but another in other ways, our evaluation of it isn’t going to depend so much on our assessment of each of these features, but on which of them we consider most salient.

I think this is part of what’s going on with current discussions of price controls. There has been a lot of heated debate following Zach Carter’s New Yorker profile of Isabella Weber on whether the energy price regulation adopted by Germany can be described as a form of price controls. Much of this criticism is clearly in bad faith. But the broad space between orthodox inflation-control policy, on the one hand, and comprehensive World War II style price ceilings, on the other, means that there is room for legitimate disagreement about how we describe policies somewhere in the middle. If you think that the defining feature of price regulation is that government is deciding how much people should pay for particular commodities, you will probably include the German policy. If you’re focused on other dimensions of it, you might not.

I am not going to say more about this topic now, though I hope to return to it in the future. But I think there is something parallel going on in the derisking debate.

People who talk about industrial policy mean some deliberate government action to shift the sectoral composition of output — to pick winners and losers, whether at the industry or firm level. But of course, there are lots of ways to do this. (Indeed, as people sometimes point out, governments are always doing this in some way — what distinguishes “industrial policy” is that it is visible effort to pick different winners.) Given the range of ways governments can conduct industrial policy, and their different implications for larger political-economy questions, it makes sense to try to distinguish different models. Daniela Gabor’s paper was a very helpful contribution to this.

The problem, again, is that models are ideal types — they identify discrete poles in a continuous landscape. We need abstractions like this — there’s no other way to talk about all the possible variation on the multiple dimensions on which we can describe real-world situations. If the classification is a good one, it will pick out ways in which variation on one dimension is linked to variation on another. But in the real world things never match up exactly; which pole a particular point is closer to will depend on which dimension we are looking at.

In our current discussions of industrial policy, four dimensions seem most important — four questions we might ask about how a government is seeking to direct investment to new areas. Here I’ll sketch them out quickly; I’ll explore them in a bit more detail below.

First is ownership — what kind of property rights are exercised over production? This is not a simple binary. We can draw a slope from for-profit private enterprises, to non-profits, to publicly-owned enterprises, to direct public provision.

Second is the form of control the government exercises over investment (assuming it is not being carried out directly by the public sector). Here the alternatives are hard rules or incentives, the latter of which can be positive (carrots) or negative (sticks).

The third question is whether the target of the intervention is investment in the sense of creation of new means of production, or investment in the sense of financing. 

The last question is how detailed or fine-grained the intervention is — how narrowly specified are the activities that we are trying to shift investment into and out of?

“Derisking” in its original sense had specific meaning, found in the upper right of the table. The idea was that in lower-income countries, the binding constraint on investment was financing. Because of limited fiscal capacity (and state capacity more generally), the public sector should not try to fill this gap directly, but rather to make projects more attractive to private finance. Offering guarantees to foreign investors would make efficient use of scarce public resources, while trusting profit motive to guide capital to socially useful projects.

In terms of my four dimensions, this combines private ownership and positive incentives with broad financial target.

The opposite case is what Daniela calls the big green state. There we have public ownership and control of production, with the state making specific decisions about production on social rather than monetary criteria. 

For the four of us on the panel, and for most people on the left, the second of these is clearly preferable to the first. In general, movement from the upper right toward the lower left is going to look like progress.

But there are lots of cases that are off the diagonal. In general, variation on each of these dimensions is independent of variation on the others. We can imagine real world cases that fall almost anywhere within the grid.

Say we want more wind and solar power and less dirty power.

We could have government build and operate new power plants and transmission lines, while buying out and shutting down old ones.

We could have a public fund or bank that would lend to green producers, along with rules that would penalize banks for holding assets linked to dirty ones.

We could have regulations that would require private producers to reduce carbon emissions, either setting broad portfolio standards or mandating the adoption of specific technologies.

Or we could have tax credits or similar incentives to encourage voluntary reductions, which again could be framed in a broad, rules-based way or incorporate specific decisions about technologies, geography, timelines, etc.

As we evaluate concrete initiatives, the hard question may not be where we place them in this grid nor on where we would like to be, but how much weight we give to each dimension. 

The neoliberal consensus was in favor of private ownership and broad, rules-based incentives, for climate policy as in other areas. A carbon price is the canonical example. For those of us on the panel, again, the consensus is  that the lower left corner is first best. But at the risk of flattening out complex views, I think the difference between let’s say Daniela on one side and Skanda Amarnath (or me) on the other is the which dimensions we prioritize. Broadly speaking, she cares more about movement in horizontal axis, as I’ve drawn the table, with a particular emphasis on staying off of the right side. While we care more about vertical axis, with a particular preference for the bottom row. 

Some people might say it doesn’t matter how you manage investment, as long as you get the clean power. But here I am completely on (what I understand to be) Daniela’s side. We can’t look at policy in isolation, but have to see it as part of a broader political economy, as part of the relationship between private capital and the state. How we achieve our goals here matters for more than the immediate outcome, it shifts the terrain on which next battle will be fought. 

But even if we agree that the test for industrial policy is whether it moves us toward a broader socialization of production, it’s not always easy to evaluate particular instances.

Let’s compare two hypothetical cases. In one, government imposes strict standards for carbon emissions, so many tons per megawatt. How producers get there is up to them, but if they don’t, there will be stiff fines for the companies and criminal penalties for their executives. In the second case, we have a set of generous tax credits. Participation is voluntary, but if the companies want the credits they have to adopt particular technologies on a specified schedule, source inputs in a specified way, etc. 

Which case is moving us more in the direction of the big green state? The second one shifts more expertise and decision making into the public sector, it expands the domain of the political not just to carbon emissions in general but to the organization of production. But unlike the first, it does not challenge the assumption that private profitability is the first requirement of any change in the organization of production. It respects capital-owners’ veto, while the first does not. 

(Neoliberals, it goes without saying, would hate both — the first damages the business climate and discourages investment, while the second distorts market more.) 

Or what about if we have a strict rule limiting the share of “dirty” assets in the portfolios of financial institutions? This is the path Europe seems to have been on, pre IRA. In our discussion, Daniela suggested that this might have been better, since it had more of an element of discipline — it involved sticks rather than just subsidy carrots. To Skanda or me, it looks weak compared with the US approach, both because it focuses on financing rather than real investment, and because it is based on a broad classification of assets rather than trying to identify key areas to push investment towards. (It was this debate that crystallized the idea in this post for me.)

Or again, suppose we have a sovereign wealth fund that takes equity stakes in green energy producers, as Labour seems to be proposing in the UK. How close is this to direct public provision of power?

In the table, under public ownership, I’ve distinguished public provision from public enterprise. The distinction I have in mind is between a service that is provided by government, by public employees, paid for out of the general budget, on the one hand; and entities that are owned by the government but are set up formally as independent enterprises, more or less self-financing, with their own governance, on the other. Nationalizing an industry, in the sense of taking ownership of the existing businesses, is not the same as providing something as a public service. To some people, the question of who owns a project is decisive. To others, a business where the government is the majority stakeholder, but which operates for profit, is not necessarily more public in a substantive sense than a business  that isprivately owned but tightly regulated.

Moving to the right, government can change the decisions of private businesses by drawing sharp lines with regulation — “you must”; “you must not” — or in a smoother way with taxes and subsidies. A preference for the latter is an important part of the neoliberal program, effectively shifting the trading -off of different social goals to the private sector; there’s a good discussion of this in Beth Popp Berman’s Thinking Like an Economist. On the other side, hard rules are easier to enforce and better for democratic accountability — everybody knows what the minimum wage is. Of course there is a gray area in between: a regulation with weak penalties can function like a tax, while a sufficiently punitive tax is effectively a regulation.

Finally, incentives can be positive or negative, subsidies or taxes. This is another point where Daniela perhaps puts more stress than I might. Carrots and sticks, after all, are ways of getting the mule to move; either way, it’s the farmer deciding which way it goes. That said, the distinction certainly matters if fiscal capacity is limited; and of course it matters to business, who will always want the carrot.

On the vertical axis, the big distinction is whether what is being targeted is investment in the sense of the creation of new means of production, or investment in the sense of financing. Let’s step back a bit and think about why this matters.

There’s a model of business decision-making that you learn in school, which is perhaps implicitly held by people with more radical politics. Investment normally has to be financed; it involves the creation of real asset and a liability, which is held somewhere in financial system. You build a $10 million wind turbine, you issue a $10 million bond. Which real investment is worth doing, then, will depend on the terms on which business can issue liabilities. The higher the interest rate on the bond, the higher must be the income from the project it finances, to make it worth issuing.

Business, in this story, will invest in anything whose expected return exceeds their cost of capital; that cost of capital in turn is set in financial markets. From this point of view, a subsidy or incentive to holders of financial assets is equivalent to one to the underlying activity. Telling the power producer “I’ll give you 10 percent of the cost of the turbine you built” and telling the bank “I’ll give you 10 percent of the value of the bond you bought” are substantively the same thing. 

As I said, this is the orthodox view. But it also implicitly underlies an analysis that talks about private capital without distinguishing between “capital” as a quantity of money in financial form, and “capital” as the concrete means of production of some private enterprise. If you don’t think that the question “what factory should I build” is essentially the same as the question “which factory’s debt should I hold?”, then it doesn’t make sense to use the same word for both.

Alternatively, we might argue that the relevant hurdle rate for private investment is well above borrowing costs and not very sensitive to them. Investment projects must pass several independent criteria and financing is often not the binding constraint. The required return is not set in financial markets; it is well above the prevailing interest rate and largely insensitive to it. If you look at survey evidence of corporate investment decisions, financing conditions seem to have very little to do with it.  If this is true, a subsidy to an activity is very different from a subsidy to financial claims against that activity. (A long-standing theme of this blog is the pervasive illusion by which a claim on an income from something is equated with the thing itself.)

Daniela defines derisking as, among other things, “the production of investibility”, which I think is exactly right as a description of the (genuine and important) trend toward endlessly broadening the range of claims that can be held in financial portfolios. But I am not convinced it is a good description of efforts to encourage functioning businesses to expand in certain directions. Even though we use the word “invest” for both.

Conversely, when financing is a constraint, as it often is for smaller businesses and households, it takes the form of being unable to access credit at all, or a hard limit on the quantity of financing available (due to limited collateral, etc.), rather than the price of it. One lesson of the Great Recession is that credit conditions matter much more for small businesses than for large ones. So to the extent that we want to work through financing, we need to be targeting our interventions at the sites where credit constraints actually bind. (The lower part of the top row, in terms of my table.) A general preference for green assets, as in Europe, will not achieve much; a program to lend specifically for, say, home retrofits might. 

This leads to the final dimension, what I am calling fine-grained versus broad or rules-based interventions. (Perhaps one could come up with better labels.) While for some people the critical question is ownership, for others — including me — the critical question is market coordination versus public coordination. It is whether we, as the government, are consciously choosing to shift production in specific ways, or whether we are setting out broad priorities and letting prices and the profit motive determine what specific form they will take. This — and this may be the central point of this post — cuts across the other criteria. Privately-owned firms can have their investment choices substantively shaped by the public. Publicly-owned firms can respond to the market. 

Or again, yes, one way of distinguishing incentives is whether they are positive or negative. But another is how precise they are — in how much detail they specify the behavior that is to be punished or rewarded. A fine-grained incentive effectively moves discretion about specific choices and tradeoffs to the entity offering the incentive. A broad incentive leaves it to the receiver. An incentive conditioned on X shifts more discretion to the public sector than an incentive conditioned on any of X, Y or Z, regardless of whether the incentive is a positive or negative. 

Let me end with a few concrete examples.

In her paper, Daniela draws a sharp distinction between the IRA and CHIPS Act, with the former as a clear example of derisking and the latter a more positive model. The basis for this is that CHIPS includes penalties and explicit mandates, while the IRA is overwhelmingly about subsidies.3. This is reflected in the table by CHIPS’ position to the left of the IRA. (Both are areas rather than points, given the range of provisions they include.) From another point of view, this is a less salient distinction; what matters is that they are both fairly fine-grained measures to redirect the investment decisions of private businesses. If you focus on the vertical axis they don’t look that different.

Similarly, Daniela points to things like the ECB’s climate action plan, which creates climate disclosure requirements for bank bond holdings and limits the use of carbon-linked bonds as collateral, as a possible alternative to the subsidy approach. It is true that these measures impose limits and penalties on the private sector, as opposed to the bottomless mimosas of the IRA. But the effectiveness of these measures would require a strong direct link from banks’ desired bond holdings, to the real investment decisions of productive businesses. I am very skeptical of such a link; I doubt measures like this will have any effect on real investment decisions at all. To me, that seems more salient.

The key point here is that Daniela and I agree 100% both that private profit should not be the condition of addressing public needs, and that the public sector does need to redirect investment toward particular ends. Where we differ, I think, is on which of those considerations is more relevant in this particular case.

If the EPA succeeds in imposing its tough new standards for greenhouse gas emissions from power plants, that will be an example of a rules-based rather than incentive-based policy. This is not exactly industrial policy — it leaves broad discretion to producers about how to meet the standards. But it is still more targeted than a carbon tax or permit, since it limits emissions at each individual plant rather than allowing producers to trade off lower emissions one place for higher emissions somewhere else.

Finally, consider the UK Labour Party’s proposal for a climate-focused National Wealth Fund, or similar proposals for green banks elsewhere. The team at Common Wealth has a very good discussion of how this could be a tool for actively redirecting credit as part of a broader green industrial policy. But other supporters of the idea stress ownership stakes as an end in itself. This is similar to the language one hears from advocates of social wealth funds: The goal is to replace private shareholders with the government, without necessarily changing anything about the companies that the shares are a claim on. 4 From this point of view, there’s a critical difference between whether the fund or bank has an equity stake in the businesses it supports or only makes loans.

To me, that doesn’t matter. The important question is does it acts as an investment fund, buying the liabilities (bonds or shares or whatever) of established business for which there’s already a market? Or does it function as more of a bank, lending directly to smaller businesses and households that otherwise might not have access to credit? This would require a form of fine-grained targeting, as opposed to buying a broad set of assets that fit some general criteria.5 Climate advocate showing to shape the NWF need to think carefully about whether it’s more important for it to get ownership stakes or for it to target its lending to credit-constrained businesses.

My goal in all this is not to say that I am right and others are wrong (though obviously I have a point of view). My goal is to try to clarify where the disagreements are. The better we understand the contours of the landscape, the easier it will be to find a route toward where we want to go.

UPDATE, MAY 2025:  Two-plus years after first publishing this post, I realize that I described Daniela Gabor’s definition of derisking as “the production of inevitability.” A poetic phrase, perhaps, but not the same as “the production of investibility,” which is what she wrote.

At Barron’s: Do Interest Rates Really Drive the Economy?

(I write a more-or-less monthly opinion piece for Barron’s. This is my contribution for March 2023; you can find the earlier ones here.)

When interest rates go up, businesses spend less on new buildings and equipment. Right?

That’s how it’s supposed to work, anyway. To be worth doing, after all, a project has to return more than the cost of financing it. Since capital expenditure is often funded with debt, the hurdle rate, or minimum return, for capital spending ought to go up and down with the interest rate. In textbook accounts of monetary policy, this is a critical step in turning rate increases into slower activity.

Real economies don’t always match the textbook, though. One problem: market interest rates don’t always follow the Federal Reserve. Another, perhaps even more serious problem, is that changes in interest rates may not matter much for capital spending.

A fascinating new study raises new doubts about how much of a role interest rates play in business investment.

To clarify the interest-investment link, Niels Gormsen and Kilian Huber — both professors at the University of Chicago Booth School of Business — did something unusual for economists. Instead of relying on economic theory, they listened to what businesses themselves say. Specifically, they (or their research assistants) went through the transcripts of thousands of earnings calls with analysts, and flagged any mention of the hurdle rate or required return on new capital projects. 

What they found was that quoted hurdle rates were consistently quite high — typically in the 15-20% range, and often higher. They also bore no relationship to current interest rates. The federal funds rate fell from 5.25% in mid-2007 to zero by the end of 2008, and remained there through 2015. But you’d never guess it from the hurdle rates reported to analysts. Required returns on new projects were sharply elevated over 2008-2011 (while the Fed’s rate was already at zero) and remained above their mid-2000s level as late as 2015. The same lack of relationship between interest rates and investment spending is found at the level of individual firms, suggesting, in Gormsen and Huber’s words, that “fluctuations in the financial cost of capital are largely irrelevant for [business] investment.”

While this picture offers a striking rejection of the conventional view of interest rates and investment spending, it’s consistent with other research on how managers make investment decisions. These typically find that changes in the interest rate play little or no role in capital spending. 

If businesses don’t look at interest rates when making investment decisions, what do they look at? The obvious answer is demand. After all, low interest rates are not much of an incentive to increase capacity if existing capacity is not being used. In practice, business investment seems to depend much more on demand growth than on the cost of capital. 

(The big exception is housing. Demand matters here too, of course, but interest rates also have a clear and direct effect, both because the ultimate buyers of the house will need a mortgage, and because builders themselves are more dependent on debt financing than most businesses are. If the Fed set the total number of housing permits to be issued across the country instead of a benchmark interest rate, the effects of routine monetary policy might not look that different.)

If business investment spending is insensitive to interest rates, but does respond to demand, that has implications for more than the transmission of monetary policy. It helps explain both why growth is so steady most of the time, and why it can abruptly stall out. 

As long as demand is growing, business investment spending won’t be very sensitive to interest rates or other prices. And that spending in turn sustains demand. When one business carries out a capital project, that creates demand for other businesses, encouraging them to expand as well. This creates further demand growth in turn, and more capital spending. This virtuous cycle helps explain why economic booms can continue in the face of all kinds of adverse shocks — including, sometimes, efforts by the Fed to cut them off.

On the other hand, once demand falls, investment spending will fall even more steeply. Then the virtuous cycle turns into a vicious one. It’s hard to convince businesses to resume capital spending when existing capacity is sitting idle. Each choice to hold back on investment, while individually rational, contributes to an environment where investment looks like a bad idea. 

This interplay between business investment and demand was an important part of Joseph Schumpeter’s theory of business cycles. It played a critical role in John Maynard Keynes’ analysis of the Great Depression. Under the label multiplier-accelerator models, it was developed by economists in the decades after World War II. (The multiplier is the link from investment to demand, while the accelerator is the link from demand growth to investment.) These theories have since fallen out of fashion among economists. But as the Gormsen and Huber study suggests, they may fit the facts better than today’s models that give decisive importance to the interest rate controlled by the Fed.

Indeed, we may have exaggerated the role played in business cycles not just of monetary policy, but of money and finance in general. The instability that matters most may be in the real economy. The Fed worries a great deal about the danger that expectations of higher inflation may become self-confirming. But expectations about real activity can also become unanchored, with even greater consequences. Just look at the “jobless recoveries” that followed each of the three pre-pandemic recessions. Weak demand remained stubbornly locked in place, even as the Fed did everything it could to reignite growth.

In the exceptionally strong post-pandemic recovery, the Fed has so far been unable to disrupt the positive feedback between rising incomes and capital spending. Despite the rate hikes, labor markets remain tighter than any time in the past 20 years, if not the past 50. Growth in nonresidential investment remains fairly strong. Housing starts have fallen sharply since rates began rising, but construction employment has not – at least not yet. The National Federation of Independent Business’s survey of small business owners gives a sharply contradictory picture. Most of the respondents describe this as a very poor moment for expansion, yet a large proportion say that they themselves plan to expand and increase hiring. Presumably at some point this gap between what business owners are saying and what they are doing is going to close – one way or the other. 

If investment responded strongly to interest rates, it might be possible for the Fed to precisely steer the economy, boosting demand a little when it’s weak, cooling it off when it gets too hot. But in a world where investment and demand respond mainly to each other, there’s less room for fine-tuning. Rather than a thermostat that can be turned up or down a degree or two, it might be closer to the truth to say that the economy has just two settings: boom and bust.

At its most recent meeting, the Fed’s forecast was for the unemployment rate to rise one point over the next year, and then stabilize. Anything is possible, of course. But in the seven decades since World War Two, there is no precedent for this. Every increase in the unemployment rate of a half a point has been followed by a substantial further rise, usually of two points or more, and a recession. (A version of this pattern is known as the Sahm rule.) Maybe we will have a soft landing this time. But it would be the first one.

 

Corporate cashflows, 1960-2016

Here is some background on the investment question from the previous post, and related topics.

I’ve been fooling around recently with assembling a comprehensive account of sources and uses of funds for the US corporate sector from the Integrated Macroeconomic Accounts (IMA). (It’s much easier to do this with the IMAs than by combining the NIPAs with the financial accounts from the Fed.) The goal is a comprehensive account of flows of money into and out of the corporate sector, grouped in a sensible way.

My goal here is not to make any specific argument, but to provide context for a bunch of different arguments about the finances of US businesses. I think this an important thing to do – both mainstream and heterodox people tend to make claims about specific sets of flows in specific periods, but it’s important to start from the overall picture. Otherwise you don’t know what questions it makes sense to ask. It’s also important to give a complete set of flows, for the same reasons and also to check that one’s claims are logically coherent. Needless to say, you also have to measure everything consistently.

Some people do do this, of course — the social accounting matrices of Lance Taylor and company are the best versions I know of. But it’s relatively rare.

The IMAs are a fairly new set of national accounts, motivated by two goals. First, to combine the “real” flows tracked by the BEA with the financial flows and balance-sheet positions tracked by the Fed into a single, consistent set of accounts; and second, to produce a set of US accounts that conform to the System of National Accounts (SNA) followed by most of the rest of the world. (The SNAs are sort of the metric system of national accounts.) The first goal is more completely realized than the second – there are some important differences between the IMAs and SNAs. For our purposes, the most important one is the definition of the corporate sector.  In the SNAs corporate businesses include, broadly, any enterprise staffed mainly by wage workers that produces goods and services for sale; this includes closely-held firms, government-owned enterprises, and many nonprofits. In the IMAs, the corporate sector is based on tax status, and so excludes partnerships and small family businesses, nonprofits, and government enterprises.

The nonfinancial corporate sector on the IMA definition accounts for roughly 50 percent of US value-added. [1] I think there are good reasons to focus on this 50 percent. This is where most important productive activity takes place, and where essentially all the profit that economic life is organized around is generated. It’s also the sector where the conceptual categories of economics best correspond to observables. We don’t directly see output in public sector or nonprofits, don’t directly see wages and profits in noncorporate sector, we don’t see either in the household sector. Finance of course has its own issues.

In any case! Figure 1 shows the corporate sector’s share of value added since 1960.

Figure 1

 

I am not sure what substantive significance, if any, most of the movements in this figure have. Some large part, perhaps most, of them reflect definitional or measurement factors rather than any change in concrete economic activity. That said, the secular rise in finance as well as government does, I think, reflect changes in what people do all day. The only one of these lines that definitely means what it seems, is the long-run rise in government – given the way the accounts are constructed, there must be a corresponding rise in the share of public sector employment. The household sector line basically reflects changes in the weight of spending associated with owner-occupied housing – the nonprofit piece of this is fairly stable over time. The fall and rise in the noncorporate business sector may also reflect the changing weight of real estate – where noncorporate forms are common – and independent-contractor arrangements. But it may also reflect shifts in legal forms and/or BEA imputations, that don’t involve any substantive change in productive activity.

Nonetheless this figure is important — less for what it tells us about economic substance than for what it tells us about economic data. Any series that exclusively or disproportionately draws from the corporate sector (nonresidential investment is an obvious and important case) will be scaled by that top line. And any discussion of factor shares needs to take into account the change in the shares of sectors where wages and/or profits are not directly observed.

Figure 2 is the real point of this post. It’s my broadest summary of sources and uses of funds in the corporate sector. All are measured as a share of total corporate value added. The same data is shown in the table at the end.

Figure 2

 

I’ve organized this in a somewhat nonstandard way, but which I think is appropriate for the questions we are most interested in. The vertical scale is fraction of corporate value-added, or output. The heavy black line shows the share of output available to corporate managers. Above the line are three deductions from value-added: first, wages and other compensation of labor; second, in gray, taxes, including both taxes on production and corporate income taxes; and third, the narrow white band, net payments to the financial system. This last is interest and other property payments, less interest, dividends and other property payments received. These are the three categories of payments that are effectively imposed on corporations from outside. [2] The area below this line is the internal funds at the disposal of management – what’s often referred to as corporate cashflow.

In red are two main uses of funds by corporate managers. The bottom red area is investment. Above this is payouts — first dividends, and then the top red area, net share repurchases. This latter includes both repurchases in the strict sense and shares retired through cash mergers and acquisitions – aggregate data combines them. The difference between the black line and the red line is net financial saving by the corporate sector. Where the heavy black line is above the top red line, the corporate sector is a net lender in financial markets – its acquisitions of financial assets are greater than the new debt it is incurring. Where the red line is above the black line, as it usually is, the corporate sector is a net borrower – its new debt is greater than its acquisition of financial assets.

Finally, the dotted black line shows reported depreciation. (Consumption of fixed capital in the jargon of the accounts.) This is not actually a source or use of funds. And there are serious conceptual and measurement issues with defining it – so much so that, in my view, it’s probably not a usable category for describing real world economies. Nonetheless, it is necessary to define some other terms that play a big part in these discussions. Most importantly, profits can be regarded as the difference between cashflow and depreciation. [3] And net investment is the difference between investment and depreciation.

The same items are presented in the table at the end of the post, for three periods and for the most recent full year available.

As I discuss below, some terms are grouped here differently from the way they are presented in the IMAs. Obviously, how exactly we aggregate is open to debate, and the pros and cons of different choices will depend on the questions we are trying to answer. But I think some picture like this has to be the starting point for any kind of historical discussion of the US economy.

So what do we see?

First, the labor share (i.e. labor costs as a percent of value added) is quite stable around 63-64 percent of value added between 1960 and 2000. It only begins falling in 2002 or so, dropping about 4 points in the early 2000s and another 3 points in the wake of the Great Recession, with a modest recovery in the past couple years. This timing is quite different from the impression most people have — what you’d get from straightforwardly looking at the wage share of GDP — of a steady long-term decline from the 1970s.

There are two reasons for this difference. First, during the 1970s and 1980s, the non-wage share of labor costs (mainly health benefits) rose quite a bit, from around 5 percent to around 10 percent of total compensation. This explains why labor cost growth did not slow during this period, even though wage growth did slow. Since healthcare prices were rising quite a bit faster than overall prices during this period, the rising share of health benefits in compensation also meant that the cost of labor to employers was also rising faster than the value of compensation to workers. [4] This factor becomes less important after the early 1990s, when the non-wage share of labor compensation flatted out.

Second, the labor share in the corporate sector is quite a bit higher than the labor share in finance and noncorporate businesses — the two sectors whose share of GDP has increased in recent decades. This means that even if there were no change in factor shares within each sector, the labor share for the economy as a whole would fall. Again, I don’t know how much of the difference in factor shares between sectors is a measurement issue, how much it reflects shifting legal forms of organization of the same kinds of activities, and how much it reflects real differences in how claims on the social product are exercised. But either way, it’s important to understand that a large part of the observed fall in the labor share over the past generation is explained, at least in an accounting sense, by this shift between sectors.

Moving on to taxes, there is also a substantial fall in this claim on corporate value-added, from 16 percent in 1960 to around 11 percent today.  But here, the decrease comes earlier, in the 1960s and 1970s – the tax share has hardly changed since 1980. (I suspect that if this figure were extended to earlier dates, there would be a large fall in the tax share in the 1950s as well.) This means that after-tax profits show a more steady long-term rise than do pre-tax profits.

I should note that “taxes” here combines two items from the IMAs — taxes on production, and taxes on profits. In the national accounts, there are good reasons to separate these — taxes on production enter into the cost of output and so have to be treated as a factor payment, while taxes on profits are not part of costs and so are treated as a transfer. This distinction is critical if we are going to calculate GDP in a consistent way, but for substantive questions it’s not so important. To government, managers and other economic actors, taxes are all mandatory payments from the corporation to the state, however they are assessed.

After taxes comes net financial payments. As defined here, this is interest, rent and net current transfers, less interest, rent and dividends received. In other words, it is net payments on the corporate sector’s existing financial assets and liabilities.  It’s represented on the figure by the white space between the thin black line and the thick black line. The first thing to notice about these net payments by corporations is that they are almost always positive and never significantly negative. In other words, over the past 56 years the corporate sector as a whole has never received more income from its financial assets than it has paid on its financial liabilities. You can see that the largest share of corporate value-added going to financial payments came in the high-interest 1980s; in most other periods the balance has been close to zero.

I’ll come back to this in a later post – a next step in this project should be precisely to unpack that white section. But the fact that the net financial income of the corporate sector is small, never positive, and shows no significant trend over time, is already enough to reject one popular story about financialization, at least in its most straightforward form. It is simply not the case that nonfinancial corporations in the aggregate have turned themselves into hedge funds – have replaced profits from operations with income from financial assets. The Greta Krippner article that seems to be  the most influential version of this claim is a perfect example of the dangers of focusing on one piece of the cashflow picture in isolation. [5] She looks at financial income received by corporations but ignores financial payments made by corporations (mostly interest in both cases). So as shown in Figure 3, she mistakes a general rise in interest rates for a change in the activities of nonfinancial businesses.

Figure 3. Because she focuses on the heavy black segment in isolation, Krippner mistakes a period of high interest rates for a reorientation of nonfinancial corporations to financial profits.

 

Returning to Figure 2: After subtracting labor costs, taxes and interest and other financial claims, we are left with the heavy red line — the share of value added available as cashflow to corporate managers. This rises from 20 percent in the 1960s to as high as 25 percent in the 1990s, to around 30 percent today. This increase in the corporate profit share (gross of depreciation, net of taxes) is one of the central facts of modern US macroeconomic history.

In the broadest terms, corporations can use cashflow in three ways. They can invest it in order to maintain or grow the business; they can distribute it to shareholders; or they can retain it for later use in some financial form. This last use can be, and often is, negative, if investment and payouts are together greater than cashflow.

Investment here includes gross capital formation, defined in the national accounts as spending on durable equipment, structures, software, research and development, and the creation of intellectual property. (The last two items have been included in the national-accounts measure of investment only since 2013.) It also includes the change in private inventories and spending on nonproduced durable assets, which I assume is almost all land. This item is listed separately in the IMAs, and it’s not obvious how to handle it: Corporate purchases of land have different macroeconomic implications than spending on new means of production, but from the point of view of the people making the investment decision there’s no major difference between money spent on a building and money spent for the land it sits on. This item is generally very small — well below 1 percent of total investment — but, like inventories, it’s highly cyclical and so plays a disproportionate role in short-run fluctuations. About a tenth of the fall in investment between 2008 and 2010, for example, was in nonproduced assets.

Somewhat surprisingly, there is no downward trend in the investment share. It was 17 percent of value added in the 1960s and 1970s, versus over 18 percent in this decade, and 19 percent in the third quarter of 2017 (the most recent available).

If investment today is, if anything, historically high as a share of corporate output, why have so many people (including me!) been arguing that weak investment is a problem? There are several reasons, though perhaps none are entirely convincing.

First, as I pointed out in the previous post, in recent years there has been an unusual divergence between investment in the corporate sector and investment in the economy as a whole. Residential investment by households remains very low by historical standards; nonresidential investment by noncorporate businesses is also low. At the same time, financial and especially noncoporate businesses always invest at lower rates than nonfinancial corporations, so the rising share of these sectors leads to lower overall investment. Second, the recovery in corporate investment is relatively recent – things looked different a few years ago. Nonfinancial corporations’ investment share fell extremely sharply in 2009, to its lowest level in 45 years, and took several years to bounce back. So when we were discussing this stuff three or four years ago, the picture looked more like a secular decline. Third — and probably most relevant for my work — while investment is relatively high as a share of corporate value added, it is quite low as a share of profits or cashflow. There is a genuine puzzle of weak investment, as long as we don’t ask “why are corporations investing less?”, but instead ask “why haven’t high profits led corporations to invest more?” Fourth, there has been a large increase in reported depreciation — from around 10 percent of value added in the 1960s to around 15 percent today. While I think for a number of reasons that this number is not really meaningful, if you take it seriously, it means that while gross investment has risen slightly, net investment has fallen a lot, to about half its level in the 1960s and 70s. Finally, if you take a strong Keynesian or Kaleckian view that it’s business investment that drives shifts in demand, then the ratios shown here are not informative about the strength or weakness of investment. The ratio of investment to output, in this view, only tells us about the size of the multiplier. To assess the strength or weakness of investment, we should instead look at the absolute increase in investment over the business cycle, which — while it’s picked up a bit in the past year — is still quite low by historical standards. I’ve made this argument myself, but I wouldn’t want to push it too far — investment is not the only source of autonomous demand.

Moving on in Figure 2: Above investment is payouts – first dividends, then net share repurchases. Here we see what you’d expect: These flows have gone up a lot. Dividends have doubled from 4.5 percent of value added in the 1960s and 1970s to 9 percent today, while net repurchases have gone from less than nothing to 6 percent (and as high as 10 percent in the 2000s.) Measured as a share of corporate cashflow rather than value added, dividends have remained stable at around 50 percent. Retained earnings as conventionally defined — profits minus dividends — have also been roughly stable as a share of value added.

Including net share repurchases with dividends is the biggest way my presentation here departs from the format of the IMAs. There, net share issuance is classed as an addition to liabilities, just like issuance of new debt. Net repurchases are the same as negative issuance — the equivalent, in the IMA framework, of paying back loans. The difference, of course, is that share repurchases have no effect on the balance sheet. This is the fundamental reason I think it makes sense to group repurchases with dividends. The flow of dividend payments is not affected by the number of shares outstanding. [6] It’s also important that market participants clearly perceive share repurchases as equivalent to dividend payments. If you read the financial press, dividends and buybacks are always treated as two forms of shareholder payouts.

Personally, I don’t have any doubts that this is the right way to look at it — today. But this is a good example of how the relations between economic and accounting categories are always somewhat slippery and can change over time. Whether net share issuance should be classed with dividends (and interest payments, etc.) as a current transfer, as I do, or whether it should be considered a financing transaction, where the standard IMA presentation puts it, depends on the way these transactions are actually used – it can’t be answered a priori. Again, I think it’s reasonably clear that, given their use today, net stock repurchases should be grouped with dividends. But in the 1950s or 1960s, treating them as financing made more sense. Also, this adjustment needs to be made consistently. If we are going to count repurchases as dividends, we have to subtract them from the headline measures of retained earnings and corporate saving. We will probably want to make an equivalent adjustment to the accounts of other sectors as well, though this poses its own set of challenges.

Another thing to consider is that we see negative issuance not only when corporations repurchase their own shares, but when shares are purchased for cash as part of mergers and acquisitions. This is not necessarily a problem. If we are just adding up payments for the sector as a whole, the two sets of flows are equivalent. On a more concrete behavioral or policy level there are important differences, but we’ll pass over those for now.

If we look at dividends alone, 2016 saw them at their highest share of corporate value-added, of profits and of cashflow since the IMAs begin in 1960; and almost certainly since the 1920s. If we measure payouts as dividends plus net share repurchases, then 2016 levels were still a bit short of the peak in the mid-2000s. Share repurchases have been a bit lower (around 5 percent of value added) in 2017 than 2016; unfortunately, the quarterly IMAs don’t have dividend data, but the financial accounts suggest that dividends have declined somewhat as well. It seems that the 2-point decline in the profit share since its 2014 peak is now beginning to be reflected in payouts to shareholders. By comparison with any period before the mid-2000s, payouts are still very high. Still, their decline over the past year seems significant – though maybe the tax bill will give them a second wind.

The final item in Figure 2 is the space between the heavy red line and heavy black line. This shows the financing gap – the net financial borrowing (if positive, with the red line above the black line) or lending (if negative) by the corporate sector. In my opinion this is a much more relevant number than corporate saving as conventionally defined. As the figure shows, nonfinancial corporations are normally net borrowers in financial markets; the brief periods of net lending are all associated with deep recessions. As the figure also makes clear, however, this specific interpretation is quite sensitive to counting share repurchases as payouts. If net equity issuance is treated as a form of financing, then the aggregate corporate sector has been mostly close to a zero balance in financial markets and has more recently been a substantial net lender. On the other hand, if we think of this gap as showing the net credit-market borrowing by the nonfinancial corporate sector — as it more or less is — then the conclusion holds regardless of how you treat stock buybacks. Either way, by this measure the recent expansion is not exceptional: As of 2016 credit-market borrowing by the corporate sector was still smaller, as a share of value-added, than it was at the high points of the 1980s, 1990s or 2000s.

The same results are shown below for three periods and for the most recent year. I won’t recap the table, it’s the same stories as above. Just to be clear, the values are the averages for the periods shown for the flows listed in the second column. So for instance labor costs accounted for an average of 63 percent of corporate value-added during 1960-1979. The first column just shows the accounting relationships between the flows.

Flow 1960-1979 1980-1999 2000-2015 2016
100 – (A) Labor costs 63 64 60 59
– (B) Taxes 15 12 12 12
– (C) Net financial payments 1 2 1 1
= (D) Internal funds (cashflow) 21 22 27 29
(E) Dividends 5 5 7 9
+ (F) Net share repurchases -1 2 4 6
= (G) Payouts 4 7 11 15
(H) Investment 17 18 18 18
– (J) Depreciation 10 13 15 15
= (K) Net investment 7 5 4 3
(G) + (H) – (D) = (I) Financing gap 0 3 2 5
(D) – (J) = (L) Profits 11 9 13 14

What do we take from all this? Again, my goal here was not to make any particular substantive claim, but to lay out some essential context for more specific arguments about corporate finances that I’ll make in the future. But it is interesting, isn’t it?

 

 

[1] Value-added is the difference between sales and the cost of material inputs. It’s the best way to measure the output of various sectors. For the economy as a whole, total value-added is identically equal to GDP.

[2] Of course corporations have some control over their wage, tax and debt-service payments. But these are not mainly decision variables for corporate management in the same way that investment and shareholder payouts are. Or at least I think it’s reasonable to so regard them.

[3] Whether they are exactly this value or only approximately depends on the profits concept being used. In any case, it’s important to keep in mind that the values of depreciation used by corporations for reporting profits to financial markets and to the tax authorities, may be quite different from the depreciation reported in the national accounts.

[4] The different behavior of prices of workers’ consumption basket and of output in general was the subject of the first substantive post on this blog, seven years ago. It’s an important topic!

[5] While I don’t agree with the claims in this article, I’m a big admirer of Krippner’s other work.

[6] The big exceptions, of course, are cases that involve all of a given corporation’s shares — IPOs and transactions that take a company private. These do respectively create and extinguish dividend flows. For this reason, when using micro data, it may make sense to use gross rather than net repurchases; but this isn’t possible with the IMA data. IPOs however are a quite small part of the overall net issuance/repurchase of shares, and I am pretty sure that firms going private are as well. Private equity might create some more serious issues here — this is something I’d like to understand better. On the other hand, the advantage of using net rather than gross repurchases is that it eliminates repurchases that are simply compensating for stock issued as part of compensation packages.

How Strong Is Business Investment, Really?

DeLong rises to defend Ben Bernanke, against claims that unconventional monetary policy in recent years has discouraged businesses from investing. Business investment is doing just fine, he says:

As I see it, the Fed’s open-market operations have produced more spending–hence higher capacity utilization–and lower interest rates–has more advantageous costs of finance–and we are supposed to believe that its policies “have hurt business investment”?!?! … As I have said before and say again, weakness in overall investment is 100% due to weakness in housing investment. Is there an argument here that QE has reduced housing investment? No. Is nonresidential fixed investment below where one would expect it to be given that the overall recovery has been disappointing and capacity utilization is not high?

As evidence, DeLong points to the fact that nonresidential investment as a share of GDP is back where it was at the last two business cycle peaks.

As it happens, I agree with DeLong that it’s hard to make a convincing case that unconventional monetary policy is holding back business investment. Arguments about the awfulness of low interest rates seem more political or ideological, based on the real or imagined interests of interest-receivers than any identifiable economic analysis. But there’s a danger of overselling the opposite case.

It is certainly true that, as a share of potential GDP, nonresidential investment is not low by historical standards. But is this the right measure to be looking at? I think not, for a couple of reasons, one relatively minor and one major. The minor reason is that the recent redefinition of investment by the BEA to include various IP spending makes historical comparisons problematic. If we define investment as the BEA did until 2013, and as businesses still do under GAAP accounting standards, the investment share of GDP remains quite low compared to previous expansions. The major reason is that it’s misleading to evaluate investment relative to (actual or potential GDP), since weak investment will itself lead to slower GDP growth. [1]

On the first point: In 2013, the BEA redefined investment to include a variety of IP-related spending, including the commercial development of movies, books, music, etc. as well as research and development. We can debate whether, conceptually, Sony making Steve Jobs is the same kind of thing as Steve Jobs and his crew making the iPhone. But it’s important to realize that the apparent strength of investment spending in recent expansions is more about the former kind of activity than the latter.  [2] More relevant for present purposes, since this kind of spending was not counted as investment — or even broken out separately, in many cases — prior to 2013, the older data are contemporary imputations. We should be skeptical of comparing today’s investment-cum-IP-and-R&D to the levels of 10 or 20 years ago, since 10 or 20 years ago it wasn’t even being measured. This means that historical comparisons are considerably more treacherous than usual. And if you count just traditional (GAAP) investment, or even traditional investment plus R&D, then investment has not, in fact, returned to its 2007 share of GDP, and remains well below long-run average levels. [3]

investment

More importantly, using potential GDP as the yardstick is misleading because potential GDP is calculated simply as a trend of actual GDP, with a heavier weight on more recent observations. By construction, it is impossible for actual GDP to remain below potential for an extended period. So the fact that the current recovery is weak by historical standards automatically pulls down potential GDP, and makes the relative performance of investment look good.

We usually think that investment spending the single most important factor in business-cycle fluctuations. If weak investment growth results in a lower overall level of economic activity, investment as a share of GDP will look higher. Conversely, an investment boom that leads to rapid growth of the economy may not show up as an especially high investment share of GDP. So to get a clear sense of the performance of business investment, its better to look at the real growth of investment spending over a full business cycle, measured in inflation-adjusted dollars, not in percent of GDP. And when we do this, we see that the investment performance of the most recent cycle is the weakest on record — even using the BEA’s newer, more generous definition of investment.

investmentcycles_broad
Real investment growth, BEA definition

The figure above shows the cumulative change in real investment spending since the previous business-cycle peak, using the current (broad) BEA definition. The next figure shows the same thing, but for the older, narrower GAAP definition. Data for both figures is taken from the aggregates published by the BEA, so it includes closely held corporations as well as publicly-traded ones. As the figures show, the most recent cycle is a clear outlier, both for the depth and duration of the fall in investment during the downturn itself, and even more for the slowness of the subsequent recovery.

investmentcycles_narrow
Real investment growth, plant and equipment only

Even using the BEA’s more generous definition, it took over 5 years for inflation-adjusted investment spending to recover its previous peak. (By the narrower GAAP definition, it took six years.) Five years after the average postwar business cycle peak, BEA investment spending had already risen 20 percent in real terms. As of the second quarter of 2015 — seven-and-a-half years after the most recent peak, and six years into the recovery — broad investment spending was up only 10 percent from its previous peak. (GAAP investment spending was up just 8.5 percent.) In the four previous postwar recoveries that lasted this long, real investment spending was up 63, 24, 56, and 21 percent respectively. So the current cycle has had less than half the investment growth of the weakest previous cycle. And it’s worth noting that the next two weakest investment performances of the ten postwar cycles came in the 1980s and the 2000s. In recent years, only the tech-boom period of the 1990s has matched the consistent investment growth of the 1950s, 1960s and 1970s.

So I don’t think it’s time to hang the “Mission Accomplished” banner up on Maiden Lane quite yet.

As DeLong says, it’s not surprising that business investment is weak given how far output is below trend. But the whole point of monetary policy is to stabilize output. For monetary policy to work, it needs to able to reliably offset lower than normal spending in other areas with stronger than normal investment spending. If after six years of extraordinarily stimulative monetary policy (and extraordinarily high corporate profits), business investment is just “where one would expect given that the overall recovery has been disappointing,” that’s a sign of failure, not success.

 

[1] Another minor issue, which I can’t discuss now, is DeLong’s choice to compare “real” (inflation-adjusted) spending to “real” GDP, rather than the more usual ratio of nominal values. Since the price index for investment goods consistent rises more slowly than the index for GDP as a whole, this makes current investment spending look higher relative to past investment spending.

[2] This IP spending is not generally counted as investment in the GAAP accounting rules followed by private businesses. As I’ve mentioned before, it’s problematic that national accounts diverge from private accounts this way. It seems to be part of a troubling trend of national accounts being colonized by economic theory.

[3] R&D spending is at least reported in financial statements, though I’m not sure how consistently. But with the other new types of IP investment — which account for the majority of it — the BEA has invented a category that doesn’t exist in business accounts at all. So the historical numbers must involve more than usual amount degree of guesswork.

Is Capital Being Reallocated to High-Tech Industries?

Readers of this blog are familiar with the “short-termism” position: Because of the rise in shareholder power, the marginal use of funds for many corporations is no longer fixed investment, but increased payouts in the form of dividends and sharebuybacks. We’re already seeing some backlash against this view; I expect we’ll be seeing lots more.

The claim on the other side is that increased payouts from established corporations are nothing to worry about, because they increase the funds available to newer firms and sectors. We are trying to explore the evidence on this empirically. In a previous post, I asked if the shareholder revolution had been followed by an increase in the share of smaller, newer firms. I concluded that it didn’t look like it. Now, in this post and the following one, we’ll look at things by industry.

In that earlier post, I focused on publicly traded corporations. I know some people don’t like this — new companies, after all, aren’t going to be publicly traded. Of course in an ideal world we would not limit this kind of analysis to public traded firms. But for the moment, this is where the data is; by their nature, publicly traded corporations are much more transparent than other kinds of businesses, so for a lot of questions that’s where you have to go. (Maybe one day I’ll get funding to purchase access to firm-level financial data for nontraded firms; but even then I doubt it would be possible to do the sort of historical analysis I’m interested in.) Anyway, it seems unlikely that the behavior of privately held corporations is radically different from publicly traded one; I have a hard time imagining a set of institutions that reliably channel funds to smaller, newer firms but stop working entirely as soon as they are listed on a stock market. And I’m getting a bit impatient with people who seem to use the possibility that things might look totally different in the part of the economy that’s hard to see, as an excuse for ignoring what’s happening in the parts we do see.

Besides, the magnitudes don’t work. Publicly traded corporations continue to account for the bulk of economic activity in the US. For example, we can compare the total assets of the nonfinancial corporate sector, including closely held corporations, with the total assets of publicly traded firms listed in the Compustat database. Over the past decade, the latter number is consistently around 90 percent of the former. Other comparisons will give somewhat different values, but no matter how you measure, the majority of corporations in the US are going to be publicly traded. Anyway, for better or worse, I’m again looking at publicly-traded firms here.

In the simplest version of the capital-reallocation story, payouts from old, declining industries are, thanks to the magic of the capital markets, used to fund investment in new, technology-intensive industries. So the obvious question is, has there in fact been a shift in investment from the old smokestack industries to the newer high-tech ones?

One problem is defining investment. The accounting rules followed by American businesses generally allow an expense to be capitalized only when it is associated with a tangible asset. R&D spending, in particular, must be treated as a current cost. The BEA, however, has since 2013 treated R&D spending, along with other forms of intellectual property production, as a form of investment. R&D does have investment-like properties; arguably it’s the most relevant form of investment for some technology-intensive sectors. But the problem with redefining investment this way is that it creates inconsistencies with the data reported by individual companies, and with other aggregate data. For one thing, if R&D is capitalized rather than expensed, then profits have to be increased by the same amount. And then some assumptions have to be made about the depreciation rate of intellectual property, resulting in a pseudo asset in the aggregate statistics that is not reported on any company’s books. I’m not sure what the best solution is. [1]

Fortunately, companies do report R&D as a separate component of expenses, so it is possible to use either definition of investment with firm-level data from Compustat. The following figure shows the share of total corporate investment, under each definition, of a group of six high-tech industries: drugs; computers; communications equipment; medical equipment; scientific equipment other electronic goods; and software and data processing. [2]

hitech

As you can see, R&D spending is very important for these industries; for the past 20 years, it has consistently exceed investment spending as traditionally defined. Using the older, narrow definition, these industries account for no greater share of investment in the US than they did 50 years ago; with R&D included, their share of total investment has more than doubled. But both measures show the high-tech share of investment peaking in the late 1990s; for the past 15 years, it has steadily declined.

Obviously, this doesn’t tell us anything about why investment has stalled in these industries since the end of the tech boom. But it does at least suggest some problems with a simple story in which financial markets reallocate capital from old industries to newer ones.

The next figure breaks out the industries within the high-tech group. Here we’re looking at the broad measure of investment, which incudes R&D.

techsectors

As you can see, the decline in high-tech investment is consistent across the high-tech sectors. While the exact timing varies, in the 1980s and 1990s all of these sectors saw a rising share of investment; in the past 15 years, none have. [3]  So we can safely say: In the universe of publicly traded corporations, the sectors we think would benefit from reallocation of capital were indeed investing heavily in the decades before 2000; but since then, they have not been. The decline in investment spending in the pharmaceutical industry — which, again, includes R&D spending on new drugs — is especially striking.

Where has investment been growing, then? Here:

hitech_oil

The red lines show broad and narrow investment for oil and gas and related industries — SICs 101-138, 291-299, and 492. Either way you measure investment, the increase over the past 15 years has dwarfed that in any other industry. Note that oil and gas, unlike the high-tech industries, is less R&D-intensive than the corporate sector as a whole. Looking only at plant and equipment, fossil fuels account for 40 percent of total corporate investment; by this measure, in some recent years, investment here has exceeded that of all manufacturing together. With R&D included, by contrast, fossil fuels account for “only” a third of US investment.

In the next post, I’ll look at the other key financial flows — cashflow from operations, shareholder payouts, and borrowing — for the tech industries, compared with corporations in general. As we’ll see, while at one point payouts were lower in these industries than elsewhere, over the past 15 years they have increased even faster than for publicly traded corporations as a whole. In the meantime:

Very few of the people talking about the dynamic way American financial markets reallocate capital have, I suspect, a clear idea of the actual reallocation that is taking place. Save for another time the question of whether this huge growth in fossil fuel extraction is a good thing for the United States or the world. (Spoiler: It’s very bad.) I think it’s hard to argue with a straight face that shareholder payouts at Apple or GE are what’s funding fracking in North Dakota.

 

[1] This seems to be part of a larger phenomenon of the official statistical agencies being pulled into the orbit of economic theory and away from business accounting practices. It seems to me that allowing the official statistics to drift away from the statistics actually used by households and businesses creates all kinds of problems.

[2] Specifically, it is SICs 83, 357, 366, 367, 382, 384, and 737. I took this specific definition from Brown, Fazzari and Petersen. It seems to be standard in the literature.

[3] Since you are probably wondering: About two-thirds of that spike in software investment around 1970 is IBM, with Xerox and Unisys accounting for most of the rest.