Digging Into the Changing Regulatory State of the Past 10 Years

Lots of exciting stuff around this post from Michael Mandel post, The Age of Regulation Started Ten Years Ago, including comments from Ezra Klein and Tim Fernholz.

This is from a paper he wrote for the Progressive Policy Institute, Reviving Jobs and Innovation, which argues “Don’t add new regulations on innovative and growing sectors during economic downturns…But despite the weakness in financial regulation, it’s a mistake to view the post-2000 years as an era of untrammeled free-market capitalism. In fact, the evidence suggests that 2000–2007, under the Bush Administration, was actually a period of rising government influence over the economy.”

Here is the graph he produces that shows an giant increase in the percent of federal regulatory agency employment as a share of private sector employment starting around 2002, with another uptick around 2008:

I wanted to take this apart and investigate it further. What has changed in our regulatory state?  I also wanted to see how the Great Recession impacted this chart – does it account for the rise in 2009?

So I went ahead and grabbed the data from Dudley and Warren’s A Decade of Growth in the Regulators’ Budget (which was the source on regulatory employment for that chart). Look at the government numbers, in 2002 there were 185,063 government regulatory workers, and in 2003 there were 242,343. In 2008, 249,392 workers, in 2010 276,429 workers. Where are these extra workers going? I took the difference of raw employment by field between 2011 (estimated) and 2000, using Table A-3 of the document above, and this is what I found (UPDATE, fixed a mistake. my chart below originally said 2010, instead of the correct measure of estimated 2011; the results are consistent for 2009, 2010, or 2011):

Homeland Security accounts for over 80% of the increase in government regulatory-designated employees during the past 11 years. Homeland Security added ~40,000 workers in Customs and Border Protection, as well as going from 0 to 60,000 workers in Transportation Security Administration. (Immigration and Customs Enforcement record 4,000 less workers.) I do not know how much of this is shuffling around people from military employment (who may not have been previously counted in this data) or whatnot.

Mandel argues that “But these agencies are not the only part of the federal regulatory sector that has grown faster than the private sector, as it turns out. The chart below shows the ratio of non-Homeland Security federal regulatory employees to private sector employment rose from 2000 to the recession year 2009.”

Over a 0.1% increase in non-Homeland Security government regulatory burden!  That looks convincing, except 2009 was a deep recession year, with unemployment over 9%.  If our economy has deep structural unemployment problems nobody told the year 2000, which was about as full-employment as we’ve ever seen lately, with unemployment at 4%.

There were simply more people working in 2000, and as such the denominator in this equation is much larger.  And government employment is sticky, so the numerator isn’t any less in 2009 – it provides services in recessions and booms. If employment drops 3%, the government doesn’t immediately fire 3% of its workers. This automatic stabilizer functionality has the added benefit of helping the macroeconomy, and it also reflects the role of government.

I reproduced this chart number two above to check that I had the numbers right. I then went ahead and, instead of using full-time equivalents, I used the definition of “labor force” by adding full-time equivalents together with the number of unemployed from the BLS. Here’s what I found:

So there’s a 0.002% growth in the non-Homeland Security regulatory burden over the past 9 years if you look at the entire labor force.

Looking at the numbers, there’s a lot of agencies gaining and losing, with a surprising number adding few if any people over the past 10 years. And what’s interesting is what other kind of regulatory agencies put on additional workers. The FDA added workers. So did the Drug Enforcement Administration, as did the Nuclear Regulatory Commission. Office of Thift Supervision went down in number of employees, which is funny as they were the go-to regulatory agency when AIG-FP and others went regulator shopping.

My absolute favorite:  One of the biggest winners over the past 9 years was the Patent and Trademark Office, which went from 6,128 employees to 10,098 employees. Given how much patents are used to shut down competition and let the largest companies rent-seek, this is probably the anti-growth part I would flag. For those who know it better, is it a symptom of court decisions? Are they playing catch-up to industry demands?

So there you have it. The Bush-era brought you a regulatory state of militarized borders, drug wars, strategically weakened financial regulatory bodies for convenient regulatory shopping, and aggressive use of patents to shut down competition. This is not the regulatory state I fight for.

This entry was posted in Uncategorized. Bookmark the permalink.

15 Responses to Digging Into the Changing Regulatory State of the Past 10 Years

  1. Rob Lyman says:

    One reason the PTO grew so much was probably the State Street Bank decision that approved business method (and expanded the potential for software) patents. There has been a flood of activity in that area, and delays at the patent office have gone way up.

  2. Kevin Capuder says:

    Thank you. As I read the opening, my first thought was Homeland Security. You saved me a serious amount of work.

  3. Pingback: Intellectual Monopoly: Patent Office: Part of the growth in the regulatory burden | Science Report | Biology News, Economics News, Computer Science News, Mathematics News, Physics News, Psychology News

  4. JCD says:


    With respect to your comments re the PTO, I wanted to offer some quick thoughts.

    To begin, the increased personnel is needed, and long overdue. The need for increased personnel is largely due to the marked increase in patent application filings over the last ten years. Without digging for hard numbers, my recollection is that the number of filings jumped from around 300,000 to around 500,000 over the last ten or so years. This is roughly consistent with the increased personnel numbers.

    The reality, however, is that the PTO under the Bush administration definitely did not hire people so they could issue more patents that could be used to shut down competition. Although it is common to hear patents portrayed as anti-competitive (I suspect this is largely due to the common association of patents with the pharmaceutical industry, as well as lobbying and marketing efforts of big technology companies), patents often enable increased competition in an industry or market. Patents can enable startups and small companies to create or enter a market without being instantly copied by larger companies with more resources. The potential for patent protection can play a large role in whether a startup or small company can obtain funding.

    Notably, the easiest way to dispose of a patent application is to allow it and let it issue as a patent. This is, coincidentally, also the most efficient strategy from a budgeting perspective, as the majority of fees the patent office receives are tied to the issuance or maintenance of a patent, rather than to the filing of a patent application.

    However, the mandate of the PTO is not to just issue everything, but rather to make sure that claimed inventions are new and nonobvious. When a case is not allowed (to issue as a patent), however, a patent applicant has a lot of different options for continuing to try to obtain the patent, thus requiring more work from the Examiner in the case. As a result, more work is typically required for patent applications that are not allowable, than for patent applications that are allowed.

    There has recently, over the past decade, been a growing political consensus that it is too easy to obtain a patent, and that such easily obtained patents are stifling the market place (and innovation). Not that the Supreme Court would ever be swayed by such political concerns, but in 2007 the legal standard of obviousness was changed owing to a Supreme Court decision. Since that time, it has become more difficult to argue that an invention is nonobvious, and thus more difficult to get a patent issued, especially in some technology areas.

    Regardless of whether these political concerns had any impact on the Court’s decision, an impact could certainly be seen on PTO policy under the last administration. In fact, many practitioners felt that, rather than striving to shut down competition as suggested, the last PTO administration seemed to be striving to shut down the PTO examination process itself. The common perception was that the PTO had adopted a mantra that rejecting a patent application was a sign of quality examination, colloquially dubbed “reject = quality”. The PTO administration proposed countless rules that seemed to be designed to make it more difficult to receive fair examination.

    One major aim of all of this was to reduce the then-growing and still outstanding backlog of pending patent applications. As noted above, the quickest way to dispose of a patent application is to allow it. But the PTO did not want to allow so many applications (as applications improperly allowed without adequate examination could stifle a market), and did not have the resources to adequately examine all of the applications it received in a timely manner. Consequently, the PTO attempted to erect new procedural hurdles allowing them to lower the number of new patent application filings they received and more quickly dispose of those they did receive (e.g. by limiting the number of continuing applications an applicant could file).

    Ironically, these efforts mostly failed (many of the proposed procedural changes were either struck down in court or postponed owing to protest and challenges and abandoned by the current PTO administration), but did result in a much lower allowance rate over the last several years. This lower allowance, rate, however, resulted in less issued patents, and thus lowered PTO revenue, as the majority of PTO revenue comes from fees associated with issued patents, not patent applications. Consequently, the PTO still lacks sufficient personnel to examine all patent application filings in a timely manner, but now is trying to recover from lowered revenues as well.

    As a broad generalization, the increase in PTO staff over the last several years can be characterized as being partly due to more resistance to issuing patents.

    In sum, the primary driver of the need for increased personnel is increased patent application filings. This need, however, has in fact been exacerbated by case law and PTO policy making it more difficult to obtain a patent, thus necessitating more PTO work to dispose of a patent application.

    Although perhaps outside the scope of your original comment, it is worth considering the composition of PTO personnel as well. Currently, the PTO has a very difficult time retaining patent examiners. The reasons for this are many, and oft debated, but can frequently be summed up as “money”. Proper examination of a patent application is ideally performed by someone well versed in the technical field of the invention the patent application is directed to. The patent office currently has very few experienced patent examiners, particularly in some technology areas, as it is very difficult to retain these individuals. Consequently, most patent applications are examined by new examiners. Professor Dennis Crouch (patentlyo.com) has found that a more experienced examiner will , on average, dispose of a patent application four months faster than a less experienced examiner. Further, while less experienced examiners frequently must consult with a supervisor, more experienced examiners will not need to consult with a supervisor nearly as much. The result of all this is that high turnover at the PTO, and a dearth of experienced examiners, causes a need for more examiners than would otherwise be needed.

    Additionally, although definitely outside the scope of your comment, it is worth noting that the PTO has a large legacy workforce that plays no role in examination. As noted by President Obama earlier this year, the “patent office receives more than 80 percent of patent applications electronically, then manually prints them out, scans them, and enters them into an outdated case management system”. Many of the examiners that I have queried about this complain that it often takes them a week to receive responses that I file electronically. Obviously, this is not an efficient use of personnel.

  5. Mike says:


    Thanks for the informed comment. I see what you are saying, the idea that more staff could mean a more skeptical stance towards patents (for better and for worse); I’ll check out some of the stuff you bring up. I suppose that’s the danger, is that the Sonny Bono patent stuff can through the legitimacy of the whole system into a cynical questioning stance, which it clearly just did for me.

  6. JCD says:


    I am glad some of my comments were worth further thought.

    Unfortunately, as with most complex systems, although practitioners and those familiar with the patent system have a lot of experience and can offer great insight, we are almost always biased, particularly the more we benefit and depend on the current system.

    The patent arena is fascinating because there are huge interests lined up on both sides. Many powerful interests would love nothing more than to never have to worry about an infringement lawsuit from a small company or individual inventor ever again, and would be willing to give up the chance for patent protection on their own inventions in exchange, while other powerful interests depend on the strength of patent protection for their very existence.

    All of that being said, I, unlike many practitioners, definitely think that it is too easy to get a patent issued for an obvious invention. The problem is that, simultaneously, it is also too common to get a novel and nonobvious invention rejected on nonsensical grounds by a patent examiner.

  7. RBR says:

    You are making the assumption that increased employment numbers in regulatory agencies is resulting in an increased regulatory burden on those regulated. Is that true, given what we know about the action and efficiency of government? Are the increased employees all regulating? making and enforcing new regulations?

  8. Ed says:

    *sad trombone*

  9. Pingback: ginandtacos.com » Blog Archive » A DIFFERENT VERSION OF THE TRUTH

  10. Pingback: Notes On Government Employment - NYTimes.com

  11. Milton says:

    Shoudn’t the first line after 3rd chart read, “Over a 0.01% increase in non-Homeland Security government regulatory burden!”? That reinforces your point more, I believe, and makes it consistent with your “0.002% growth” citation. Great post, though, sorry to be nit-picky.

  12. Bilo says:

    Also, if there are more, say meat inspectors hired, there may not be any more new regulations, but more businesses might have to cope with those pesky inspectors. Which would be disliked by some of those businesses. And if they have to change how they do things, it might cost them money, and be, as they say, “bad for business.”

    But it might be better for society. So the usual bellyachers can blow it out their asses. (Provided they’re not already doing so because of e. coli.)

  13. Pingback: Homeland Security and the Regulatory Burden « Mandel on Innovation and Growth

  14. Pingback: Mandel Responds on Innovation, Growth and the Regulatory State. « Rortybomb

  15. Pingback: The burden of regulation « Phil Ebersole's Blog

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s