by Guy Rolnik at Pro Market blog.
"Looking at both intangible investments and political activities to explain the 20 percent rise in Tobin’s q in
the U.S. since 1970, a new working paper by James Bessen from Boston
University concludes that activity associated with increased federal
regulation is the most important explanatory factor, especially after
2000. In fact, spending on R&D and other intangibles has fallen,
relative to conventional assets, since 2000.
Noting that operating margins for these
firms have also risen since 1990 by over 2 percent in aggregate,
Bessen’s study also found that variables associated with regulation and
corporate campaign contributions account for about half of this
increase.
When expressing the political rent
seeking effect in dollar terms, the paper states that it corresponds to
an increase in the value of non-financial public corporations of about
$2 trillion, and that this amounts to an annual transfer from consumers
to firms of about $200 billion. Several tests performed by Bessen also
suggest that the link between industry regulation and corporate profits
is indeed causal, flowing from regulation to profits.
The RegData technique provides deep
insight into the complexity of regulation. For example, last January,
the Mercatus Center at George Mason University, which hosts the RegData
project, published “The McLaughlin-Sherouse List: The 10 Most-Regulated Industries of 2014,”which
was led by petroleum and coal products manufacturing (25,482
restrictive words), electrical power generation, transmission and
distribution (20,959 restrictive words), and motor vehicle manufacturing
(16,757 restrictive words). Bessen used RegData
to gain a deeper understanding of the relationship between regulation,
rent seeking and the balance of power between corporations and
consumers.
Bessen’s paper corresponds with a 2015 presentation by Jason Furman and Peter Orszag,
in which they explored the potential role of rents in the rise in
inequality, and in allowing firms to achieve super-normal returns. After
comparing the distribution of returns on equity across S&P 500
firms between 1996 and 2014, they found a distinct possibility of “an
increased prevalence of super-normal returns over time” (which the
distribution skewed to the high-end over time, since the firm at the
high-end of the distribution earn more super-normal returns). After
reviewing Census Bureau data on market consolidation showing that in
three quarters of the industry covered by the census, the 50 largest
firms have experienced a rise in revenue share, the two concluded that
“consolidation may be contributing to the changing distribution of
capital returns and the increased share of firms with apparently
super-normal returns.”
Bessen’s path to his current academic post, a lecturer at the Boston University School of Law, was unusual.
Bessen studied economics at Harvard.
Later he became a software engineer and developed the first commercially
successful “what-you-see-is-what-you-get” (WYSIWYG) PC publishing
program. In 1993 he sold the company he founded, Bestinfo. He returned
to academia, working with Nobel Laureate Eric Maskin on a paper about
software patents. In 2002 he became a visiting scholar at MIT’s Sloan
School of Management, and in 2004 he joined Boston University.
James Bessen: My main research
focus is technology. I have a technology background. I ran a software
company and it’s what I have been studying for about 15 to 20 years.
This is my second piece on political economy. The first was that Foreign
Affairs piece.
As I was looking at what was happening in
technology policy across many different areas, I saw the same thing
happening, whether you are talking about trade secret law, patents,
non-compete agreements for employees, or government procurement policy.
It just seems that policy is working against startups in way not seen in
the past. It seemed to be in favor of the incumbent firms. That
observation led me to the Foreign Affairs piece.
GR: Did it come from your academic work, or was it something that you picked up when you were still in industry?
JB: It came more from the academic
work. A lot of my work was in the area of patents. When I was in
industry, I was in the software business. Just after I sold my company,
they started issuing software patents in large numbers. Immediately,
people in the software industry were upset by this.
They felt that, “Here’s a very innovative
industry. Why do you want to introduce patents and lawyers into
something that’s working very well without them?” To this day, most
software developers are opposed to patents on software, or at least most
patents on software.
I started a line of research that looked
at that controversy. What happened was, it turned out that it was the
well-established hardware companies, and then later on, the large
software companies, were in favor of these patents, but most software
developers were not.
Even as recently as four or five years ago, start-up firms generally didn’t get patents in software.
These patents benefit large hardware and
software companies, sometimes at the expense of startups. But because
this change created large numbers of software patents with poorly
defined property boundaries, patent litigation started to increase
sharply. Many of these lawsuits were filed by patent trolls who used
lawsuits to extract settlement payments from startups.
As we did more empirical research on what
was going on with patents and patent trolls, and it became a
legislative issue, it was very striking.
We did a lot of work on patent trolls
that was entered into the debate as firms were pushing for legislative
reform on patents. The lobbying that went on was so striking. It was
huge. It was an education for me.
Where I’d been thinking in very academic
terms, there is this brutal policy landscape and conflict that goes on.
That has a whole lot to do with how this policy has evolved in the
beginning and how it is evolving today.
GR: When you say it was an
educating experience for you, do you mean that you realized that
regulation and patents work for the big incumbents?
JB: Right. Not only do they work for them, but that the political process was centered around them.
I’ll give you a little backstory. There
was a lot of discontent that started bubbling up in the late 2000s, a
huge increase in lawsuits.
Lots of small companies were sued both in
tech fields and even outside of tech. A lot of retailers, a lot of
travel or hotel industry people, who all of a sudden are upset about
patents because they’re getting sued over this whole range of patents
that should never have been issued, but it becomes a legal strategy to
extract money out of them.
In 2011 a new patent law passed, the
Leahy-Smith America Invents Act. This patent law was essentially
negotiated between a small number of large pharma companies and a small
number of large tech companies.
This law has some good things about it,
but there was a huge amount of lobbying, a huge amount of money that
poured into this, and they came up with something that was very tame.
Because it didn’t address the problems,
all of a sudden you have a whole lot of small businesses in every state
in the country who are now upset about getting sued for patent
infringement over these very ridiculous claims.
What you saw from 2013 is this much
broader grassroots activity pushing for patent reform by a lot of
start-ups, by a lot of small companies, throughout the country.
It’s been pretty much blocked by
pharmaceutical companies, patent trial lawyers and some well-established
interests. This was my original education on how things are playing
out. It’s happening in other policy areas.
GR: What brought you to this recent paper?
JB: I’ve been reading some of the papers of Jason Furman and Peter Orszag.
GR: Do you mean the paper about the “super firms”?
These investments don’t show up in normal
accounting, in the normal balance sheet, as assets, and this has been a
well-known problem for some time. When I started out I had the
expectation that I was going to find that there were some political rent
seeking aspects to the rise in profits and the rise in corporate
valuations, but that it was probably more an issue of unmeasured
technology investments.
What I found was that I was wrong in my
prior opinion, especially for the last decade, it’s been more a story
about political rent, especially concentrated in regulated industries.
GR: How long was the period you’ve been looking at?
JB: The total period I looked at was from 1970 to 2014, but when I say, “The last decade,” I really mean from 2000 on. 2000—that
was when the tech bubble burst, the Internet bubble, and firms had been
spending a lot on technology. They cut back some of that spending.
Since then, you’ve seen a relative decrease in technology spending, but
regulation and campaign spending have gone up.
GR: Why do you think that lobbying expenditure is a good measure of rent seeking?
JB: I don’t think lobbying
expenditure itself is a very good measure. One of the things that comes
out of this paper is the suggestion that lobbying expenditure and
campaign spending are tip-of-the-iceberg measures.
What it finds is that the regulation index is the strongest factor.
GR: The RegData index.
JB: The RegData, yeah. That is
picking up. My interpretation is that when you see an industry with a
very high RegData index, a very high level of complexity or restrictive
words in the regulation, that reflects a lot of industry rent seeking
activity. But the nature of that activity is very varied and involves a
lot of different things.
It includes lobbying and campaign expenditure, but those two things are probably a small part of it.
GR: Is this a more of a Stiglerian phenomenon, or more of a tollbooth phenomenon?
JB: It’s very much a Stiglerian
type phenomenon. The tollbooth idea implies that regulators want to
extract rents from companies, and so you would expect highly regulated
industries to have decreased rents.
The Stiglerian idea is that regulation
comes about because of cooperation between regulators and industry, and
so you would see greater rents. That’s what we see. It may not be a
simple story of industry activity capturing individual regulators.
There may be a much more complex story.
It may be, for instance, that regulatory complexity helps established
firms extract rents and complexity serves as a barrier to entry to new
startups; these barriers can increase rents and corporate valuations.
GR: So the important empirical
evidence is derived from RegData correlating with the rents that are
manifested by the Tobin outcomes?
JB: Right. I’m seeing a large correlation, and that seems to be accounting for a very substantial part of the increase in rents.
It could be that we’re seeing a spurious
correlation and that perhaps it’s high-profit industries that are
attracting regulation.
At the turn of the 20th century, trusts
were these high rent, high monopoly industries that were getting
attacked by regulators because they were making high profits.
I did this causal analysis and found that
the causality really does flow the other way: when you see an increase
in regulation, profits follow.
GR: Another surprising finding of
yours is that you feel that most of the rents are derived from complex
regulations, and not from market power driven by concentration.
JB: I don’t think these things are
necessarily in opposition. The problem is that my data doesn’t speak to
it. I do use the share of the top four firms in an industry and find
that it does not have much explanatory power. But that’s only one
measurement of market power.
There may be many other sorts and it may
be that the relevant sorts of market power aren’t picked up by a simple
index like that. It may be that, for instance, when you think about the
electric utility industry, the relative market power is not the national
market but the local market, so you’re not going to pick it up there.
When you think about the pharmaceutical
industry, you may be talking about a highly differentiated product
market, so it’s not the number of pharmaceutical firms in general. It
may be the ones that are particularly involved in producing Statins, or
cancer drugs or other treatments.
There are these niche markets that they compete in and they may very well work as monopolies.
GR: Some of the industries
that you do mention that have rising regulatory complexity are also
quite concentrated or have high HHI values, like telecom, mobile, cable,
big pharma, and of course airlines, railways, and so on.
JB: That’s right. The study raises
more questions than it answers. These are some of the interesting
questions. What it speaks to is that we really don’t have our arms
around what’s going on in all these industries. We can speculate.
It’s this notion that we can look at
these concentration ratios, and they’re indicative for some industries,
but in other industries it’s not picking it up. It’s just not the best
measure.
GR: How is it that the financial sector does not appear among the five industries that you mentioned?
JB: Good question. It’s because the way that balance sheets are constructed and…
GR: Because of the Tobin’s measure, you can’t use Tobin
for financial companies. Since RegData is the base of your empirics, can
you say something about the way it works?
JB: It’s a very clever approach
they’ve got. They’re continuing to refine it. Previous measures have
just looked at the gross number of words in regulations. The developers
of RegData do two things. One is they look at restrictive words, words
like “shall,” or “must,” to rank the restrictiveness of each section of
the Federal Code. They also look at key words that relate to a
particular industry, using a sophisticated textual analysis, and that’s
where we get the real action in this case. They’re able to go and do an
annual measure, for most industries, that tells us the regulatory
complexity. It’s fascinating to look at some of these.
You see something like major legislation
gets passed—and then the regulatory complexity goes up. You see some
cases where there’s deregulation occasionally and the complexity goes
down for a period of time, but the overall trends are up.
GR: Would it be correct to say that, in a way, what
you’re saying is that this paper adds to the literature that proves that
George Stigler was right, and it’s happening in a big way in the last
15 years?
JB: Yes. The surprise here is the
magnitude of things. That’s what caught me and surprised me. I’m seeing a
very strong effect and an economically large effect. That makes me
stop, think and say “This is important stuff.” It’s Stigler being right
in a big way.
GR: In your HBR paper you say that basically lobbyists
and lobbyism are only the tip of the iceberg. Can you describe what you
think goes beneath the surface level in this?
JB: First of all, it’s not only
U.S. activity. It’s global. Second, lobbying is largely directed at
getting legislation and regulatory decisions, but there’s a lot of what
we would call rent seeking activity.
If you look at the electrical industry or
cable television, there are rates that get set. There are these complex
rules, and firms have to meet with regulators on an on-going basis.
That’s not lobbying, per se.
They may challenge those rates and those
rules in court. They need to be monitoring them. They may change their
offerings, their products, to take advantage of changes in regulation.
That was part of the story with the Cable Act of 1992 where they made
these changes.
This is the point that Richard Posner
made back in 1975. Using the example of airline regulations, he saw that
rent seeking involves actually changing the nature of their products
and services. You’ve also got activities like clinical drug trials which
are huge investments that pharmaceutical firms make with the aim of
getting regulatory approval.
All of these sorts of things are termed
as rent seeking activitis—activities where the firm invests in getting a
political outcome that increases their rent."