Arnold Kling on Knowledge, Power, and Unchecked and Unbalanced
Sep 6 2010

Arnold Kling of EconLog and author of Unchecked and Unbalanced, talks with EconTalk host Russ Roberts about the book and the relationship between knowledge and power. In a modern economy, specialization has increased and knowledge is increasingly dispersed. But political power has become more concentrated and fails to exploit the potential for decentralization. Kling discusses these trends and the potential for decentralization of power under different policies.

Bruce Bueno de Mesquita on the Political Economy of Power
Russ Roberts talks with Hoover Institution and NYU political scientist Bruce Bueno de Mesquita about his theory of political power--how dictators and democratically elected leaders respond to the political forces that keep them in office. This lengthy and intense conversation...
Arnold Kling on the Three Languages of Politics, Revisited
Economist and author Arnold Kling talks about the revised edition of his book The Three Languages of Politics in front of a live audience at the Cato Institute, recorded in September of 2019. Kling talks about the changed political landscape...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Robert Kennedy
Sep 6 2010 at 7:50pm

Kling offers a hopeful thought at the end of the podcast that “the elites” now understand that based on the experiences of the 70s that wage and price controls don’t work. I would agree that no one since suggests explicit wage and price controls anymore. On the other hand, those that believe centralized management is best are still advocating implicit wage and price controls via tariffs, quotas, trade restrictions, tax incentives, stimulus packages, market regulations, and monetary policies, all of which control wages and prices in more insidious ways. If anything, there seems to be more of a call for central planning of the economy, not less.

I’m grateful for Kling’s attempt to end on a hopeful note. I’m just not yet convinced..

Ralph Buchanan
Sep 6 2010 at 10:40pm

“It may seem strange that so many people of great intellect have said and done so many things whose consequences ranged from counterproductive to catastrophic. Yet it is not so surprising when we consider whether anybody has ever had the range of knowledge required to make the sweeping kinds of decisions that so many intellectuals are prone to make, especially when they pay no price for being wrong.

Intellectuals and their followers have often been overly impressed by the fact that intellectuals tend, on average, to have more knowledge than other individuals in their society. What they have overlooked is that intellectuals have far less knowledge than the total knowledge possessed by the millions of other people whom they disdain and whose decisions they seek to override. ”
-Thomas Sowell

Remember during the last few presidential elections, some specific candidates were advertised as ‘genius’ and great hoopla was made of their educational lineage. The omniscient enlightened despot is a political brand encouraged by some.

Home-schooling involves a home curriculum, as well as local groups (Classical Conversations for example) that take advantage of the variety of expertise offered by the parents.

I was taught as a child to treat eggs and all raw meat as if it were a potential source of disease/infection – you have to cook it and wash your hands when handling it. I’m not sure why the government recall took place – cook your food and wash your hands. If you order a rare burger or steak in a restaurant nowadays, they either refuse, or warn you that it could make you sick.
Thanks for the podcast. Kling is always great!
Congrats on the tub fix; plumbing is never fun!

Sep 7 2010 at 2:00pm

I wonder how much increased specialization will lead to increased structural unemployment. It seems that the more specialized our set of skills need to be in order to be successful in the more complex and desirable positions, the less ability we would have to adjust to changes in labor demand.

If this is true, perhaps increased specialization will make any one particular position such a small part of the labor market, that a reduction in demand for that position will have minimal macroeconomic effects.

It would also seem that maybe the labor market as a whole is becoming more diversified, while the individuals that make up the labor market are becoming less diversified.

Any thoughts?

Sep 7 2010 at 3:05pm

Will –

In regard to your comment:

It seems that the more specialized our set of skills need to be in order to be successful in the more complex and desirable positions, the less ability we would have to adjust to changes in labor demand

I would tend to agree with this sentiment and have heard a similar idea discussed many times with respect even to the current economic woes of the nation. Similar to how the internet has made dissemination of mass information easier (fewer barriers to entry), high specialization increases the barriers to entry for any labor position. As jobs that can be completed with a few hours or days of training are moved overseas, or automated and made unnecessary, replacing them with jobs that require months or years of training will likely drastically affect the economy, and likely has already. For example, states like Michigan with rapidly decreasing manufacturing employment have had high unemployment for longer than the current recession.

This may be a problem to be solved by either the educational opportunities or by individual redefinition. Either people need to be trained in a broader base of skills before entry into any employment, or they need to recognize that they already are. In the first case, a loss of jobs in one sector is less devastating because each multi-sector trained worker can move into another position with some ease. In the second case, as many resume writers suggest, people should consider their membership in clubs with leadership positions, their involvement as a leader on a project, etc. as a marketable skill and think of ways that may apply to other areas of the economy.

Regardless of specialization, all business is about trust (I believe one of my college professors said that, not to claim this level of wisdom on my own). Develop a high level of trust in professional relationships; keep an open mind in business endeavors; work hard to develop the skills required for what you want to do. Some of the rest is luck, but more often than not, that combination will work to find something, even if specialized skills are still required along the way.

Sep 7 2010 at 3:14pm

One side comment – those folks who are interested in convincing the government it may not be the best at regulating food safety and that some other system may do it better may find an ally in the local food movement (see page 2).

Sep 7 2010 at 4:52pm


‘negro’, really? It’s 2010 man.

Sep 7 2010 at 5:54pm

“For the elites, gradually losing some of the illusions they have.” -AK

Ralph provides a nice Sowell quote as to why this won’t happen. Kling says it too:

“Ability of the human mind to tune out negative feedback is tremendous; confirmation bias is great.”

I think this topic deserves more discussion. That’s one of a great weakness of the “elites”. I’ve seen it firsthand. They can rationalize away negative feedback, they have a strong tendency for confirmation bias and they don’t often suffer the direct consequences for their bad calls.

These guys need a 12 step program to get them to the point that they can readily admit that they might be wrong about something.

John Berg
Sep 7 2010 at 7:57pm

Was I mistaken? Did I hear two “elites” admit that they could not control the education of their children in Montgomery County, MD? I haven’t lived there since 1981 but haven’t they replaced every every teacher from K to 12 with one who has a Ph.D. and provided a class size of 10? Did the two throw up their hands in defeat at the power of the teacher union that controls their children’s education?

Significantly, this week marks the reaching of 50% for union members being government workers, among whom are teachers.

John Berg

PS. These podcasts are among my joys.


Robert Kennedy
Sep 7 2010 at 8:38pm

I do think Montgomery county, Maryland might be a bit of an anomaly, at least compared to other states. I’ve lived in 6 states. In all of those states other than Maryland, county government is mostly powerless. In Maine & New Hampshire, for example, counties only exist to run courts and keep records. They have nothing to do with schools or law enforcement or anything else. Montgomery County in Maryland was far & away the most powerful county government that I’ve encountered. Heck, the county even controls liquor distribution.

I agree in concept that further localization of control makes sense in theory. But in practice, I’m not so sure. We have smallish school districts in Maine, mostly organized by town boundaries and they are no less susceptible to union influence than larger districts with the further disadvantages of a lot of redundant administrators. That said, the state of Maine imposes many of the rules & regs that local school districts have to follow so I’m sort of contradicting myself. We have some local control over budgets but only some because of state mandates and the union influence. As citizens we get to vote on the school budgets but rarely do we push back because who is willing to speak out against education or teachers, huh?

Can others comment about school quality in states that don’t allow public worker collective bargaining? My gut tells me that there would higher quality lower cost education without union influence but i don’t have any empirical data to back up that notion.

Sep 7 2010 at 10:35pm

Arnold was partly wrong about Dodd-Frank stripping power from the ratings agencies. The bank regulators are not allowed to use them in their regulations. See, e.g.

Steve Bacharach
Sep 8 2010 at 11:19am

The day after hearing this podcast, I listened to a PBS Newshour Podcast titled “Gov2.0: Aims to Make Government More User-Friendly”.

It seemed to pick up exactly on some of the themes presented by Russ and Arnold. I’m sorry I can’t find a link now, but I would urge Russ, though I’m sure he’s quite busy, to give it a listen. An interview could be a great juxtaposition to the one with Arnold Kling.

Thomas A. Coss, RN
Sep 8 2010 at 1:00pm

The autocracy of healthcare is a great example at a highly micro level, what happens at the macro level withing governments.

In healthcare when the physician makes a mistake, the patient pays, both in a pecuniary and often physical way. The patient can not outsource the healing or the dying. In healthcare we still take profound umbrage when the actions of a physician result in the death of his or her patient, this is less obvious in government.

We can be grateful that we don’t use the same entry criteria for the practice of medicine that we do for the practice of government. Who here would be willing to be operated upon by a physician based upon his or her seniority, tenure or political connections?

Certainly makes one wonder about government involvement in healthcare doesn’t it?

Sep 9 2010 at 12:36am

Increasing specialization is plainly evident in some fields (academia, medicine, tax law), but is this pervasive? My view is that complexity has increased, but the role of humans in this process is problematic.

I submit that augmentation is a considerably greater factor, and one consequence is to minimize specialization. Compare the building and herding of industrial robots to the mental and physical skills required to produce an automobile before the assembly line was created. Compare building, operating and maintaining a modern container vessel to the same for an East India Company sailing vessel circa 1850.

On the whole, the trajectory seems more toward loss of specialized knowledge acquired, exercised and transmitted by individuals, and of physical skills gained by lengthy training and practice. (It’s all too easy to ignore the physical human body but it ain’t going away.) The trend is toward generic, interchangeable “human resources.”

The myopic focus on Montgomery County is puzzling. Wyoming, Vermont, North Dakota, Alaska, South Dakota, and Delaware all have populations less than that of Montgomery County. No evidence is offered to compel a listener to believe that Montgomery County, Maryland is in any way representative of the forces discussed. If the example were at least a bit removed (say, suburbs of Baltimore, Mexico City or Dacca), this listener would be more inclined to suspend disbelief.

When the conversation turned to the Swiss example, I had some hope it would become interesting. But no.

If this were a radio call-in show, I might have asked Russ and Arnold to comment on this observation by G. K. Chesterson: “Too much capitalism does not mean too many capitalists, but too few capitalists.”

John Berg
Sep 9 2010 at 1:51pm

Montgomery County, MD along with its contiguous neighboring counties in MD and VA, is atypical of most of the US, in being the home of a preponderance of “technocrats” employed by the Federal Government either directly or through contracts, grants, and employment to influence government. For example, parents and teachers can cooperate to influence money from the Feds to support education because most of the senior staff in Dept of Ed. live in these areas.

John Berg

Sep 16 2010 at 7:33pm

One of he key arguments that AK gives is the 9 council members in his local county that spend $500Million each every year. Yes, sounds like a lot, until you have spent that much money and then its not that hard.

By comparison, Wikipedia states that Forex Trading is about $4trillion a day. That 18% of this happens by Deutsche Bank, a company that has 80,000 people working for it , which works out (if you divide it equally) to approximately 80,000 people spending $3.2 billion each, every year.

These are people we don’t know. These are people you would never recognize on any street, or in the newspaper, these are people that don’t even know each other to any substantial degree even though they all have similar spending responsibilities in the same company.

What seems to be the case, is we get all romantic about Deutsche Bank, when we think of it as ‘part of the market’ and then all critical of those same actions when they are ‘part of government’. Same actions, same results, only difference is what group we put them in and we change our opinion.

Personally, I don’t see Government getting stronger. They are weaker today than they have ever been.

As an individual, I have a million choices, any of which I can make at the drop of a hat. Including ‘leaving the city, county, state, country’. What I have to do to ‘leave even the Country’ is minimal compared to what my ancestors had to do when they came to America. And so, if the Government does keep me and all my neighbors ‘happy enough to stay’, then they lose many of us.

The ‘voice’ I have in politics trumps anything seen in any past generation. The fact that I’m responding to this podcast at a whim, or that I can ‘tweat’ to a million people without chopping down a single tree, or email a million people for about a weeks wage, is something generations before us never dreamed about.

As for the taxes I pay, I don’t see this making the government any stronger. Every dollar I spend the government spends even more just trying to keep me happy. They cant even charge me what it costs to keep me happy, how powerful is that? If they were truly powerful, they would be charging more than it costs to provide their services, not less.

Even the ‘war machine’ is all but totally broken. 6 generations back, and all eternity before that, a mandatory draft was an easy thing to do. What generation didn’t have a brother, a cousin or an uncle that had died in a war? Today, I think there are a few in my city, but they were all volunteers, and I don’t know any of them.

We can get all romantic about the past, but who would trade it for the present?

Are the poor still slaves to poverty? Yes, as they have always been. But again, for a few weeks wage at the local walmart, you can buy a plane ticket to anywhere in the world.


Sep 17 2010 at 8:48pm

After listening to podcast on Kling “Unchecked Power” and Munger “Public rent seeking” and Pink “Incentives” the recent news of the $800k Bell city manager,0,7803374.story, out-earning the US president.

is an example of the lack of checks and balance in the system. Where did our system not catch this? Who provides oversight of county and administration (e.g. school) salaries, and how effectively tax payer money is allocated to provide the service vs the administrative overhead? Can you consider interviewing someone in the public sector who knows the inside of this?

Sep 20 2010 at 4:48pm

There are a few comments I would like to add to the picture painted of Switzerland.
Not only are the Kantons much smaller with as little as 30 000 inhabitants and a few miles in diameter (so that it is easy to move from one Kanton to the next, if you dont like public schools, taxes and so on) but also a substantial part of gouvernment is part-time gouvernment, so that politicians are still working in their normal work-place. Even the Swiss central parliament (the “Bundesrat”) meets several times for a few weeks for legislation, while the members remain in their jobs as entrepreneurs, lawyers, doctors, artisans and so on.
In addition, we have some kind of committee gouvernance compared to the internet – if some people see a problem, they can write a change to the law (either writing the exact paragraphs or writing a guideline how the law should be changed), then collect signatures (depending on the size of the region affected – if it is a general law they require more than 100 000 signatures, in smaller cantons or in communities it might be only a few hundered) and a public vote on this law has to be performed. For this vote, the legislator or the gouvernment often discuss a counter proposal, which could be less strict than the proposed law or with some practical changes. This counter proposal is often modified in the run up to the vote and it might even be, that the initiators withdraw their original idea and stick to the changed suggestion. If the parliament and the originators agree on one idea, the public vote is often not necessary – in any other case the general public will decide on any changes. The changes influenced by these laws can have a wide variety of subjects -from changes to a street in a community or the building of a new sports-stadium, up to changes in the tax code or even major changes in the constitution (one group was trying to reinstall the death penalty but withdrew before the signatures were collected).
The committees writing these proposals often only form to change one law. They might be supported by a party, but they also might be supported mainly by one engaged citizen.
Although the central gouvernment tries to increase its power even in Switzerland, the checks are rather high here, not at least due to the large number of small Kantons that counterbalance each other.

Comments are closed.


About this week's guest:

About ideas and people mentioned in this podcast:Articles:

Podcasts and Blogs:



Podcast Episode Highlights
0:36Intro. [Recording date: September 1, 2010.] What is the discrepancy between knowledge and power that is the focus of your book? Idea is that knowledge has become increasingly specialized and dispersed, but power is becoming increasingly concentrated. That's the discrepancy. Why is that important? We've discussed many times on this program: markets allow for dispersed knowledge to be aggregated. This is Hayek's story: information flows into prices and markets work that way. If knowledge is becoming increasingly dispersed, would suggest that you would want to have more social transactions mediated by markets. Have this oddity that political power is becoming more concentrated, perhaps having more transactions mediated by central planning. If true, that discrepancy should be causing inefficiency and general dissatisfaction. Dispersion of knowledge and observations on specialization. You confess this is not an easy thing to measure. No Gini coefficient. A lot of suggestions in the world around us. Some indicators. A Gini coefficient for knowledge--you'd think you could almost develop that. If you really decided that's what you wanted to do--some field like library science, would be cool thing to do. A lot of indications. You can track the number of medical specialties, legal specialties. College campus: many more possible majors than 30 years ago. Produce diversity has gotten higher: people argue that's a welfare increase, more different products. Some say it's a welfare decrease--can't make these choices. A lot of occupations that didn't exist years ago: friend two college graduate daughters, one a web designer for a financial firm, other works with a company that does social media marketing. Didn't exist when they were born. Proliferation of new occupations. Harder to be a Renaissance person. It was imaginable 400 years ago that you could read master a relatively large part of the world's knowledge. Seen it argued that Leibniz knew just about everything that could be known at that time! Da Vinci superior in many fields. Now a Renaissance person if you are good in a couple of things, if you know something about a lot of things. Master a couple of things. Could argue that's simply a change in the returns to mastery: the cost of mastering a lot of things has gone up. Return to specialization has gone up. Medicine, law, mathematics--the Leibniz today. How much could a Newton or Leibniz today master? Not all of it. When we went to graduate school in economics: the amount of economics you would not cover would be a lot higher today. Finance was around, but behavioral economics and experimental economics were not, or were less prominent. I used to call myself a macroeconomist--I can't follow macroeconomics; sort of can; highly mathematical, Euler equation stuff. Probably a difference between being a pediatric kidney specialist and being some kind of flavor of macroeconomist. Profusion of journals in every discipline. If you want to call yourself a master of any field, not sure that leads to true mastery. In economics, for example.
7:38Let's take as true: there's been an explosion of knowledge and a specialization within that wider span of knowledge so that any one--combination. Returns to mastery in any sub-discipline have gone up. Knowledge is more dispersed. Not just an issue for people trying to become academics. True for somebody in business: if you want to be a CEO, there are many more things you have to understand than you used to. You didn't have to understand information technology to be a CEO. Didn't have to be an expert in global supply chain management. Didn't necessarily have to understand logistics, be as sophisticated in finance. Even for consumers, amount of knowledge you need is more. More different financial instruments available for you to either trip up on or take advantage of. All sorts of different products and services that didn't exist. Economy faced with this reality, which it's creating in an emergent, Hayekian way, which you point out in the book, outsourcing, and not just overseas--less vertical integration, delegation of tasks outside the umbrella of the firm itself. For the consumer as well. Households are outsourcing more of the food preparation, cleaning, lawn care, more complex forms of electronic communication--have cell phone, do you still keep your land line? Life more complex for everybody. Weird thing. Have computer network in house, access Internet wirelessly; running an IT center. Measurement question: has my house gotten more specialized or less specialized? What am I doing with an IT network in my house? Verizon supplies it; I don't master much of it other than opening the door to let someone in to drill in my wall. Hard to measure these phenomena. Makes us very much dependent on other people's expertise. When something goes down on your network, you have to call a service person, really grateful when somebody can solve your problem. Relying on all sorts of expertise; when you buy your computer you trust it will work with your network. Scheduled for next week's podcast: psychological impact of the fact that our jobs are much narrower than they used to be. Dan Pink podcast: people needing to feel sense of mastery, source of dissatisfaction: for a lot of people they are going to have to get some of that feeling from hobbies rather than their job. See it more as a personality issue. There are a lot of people who love doing one thing, mastering it. Guy who comes to my house and sets up my wireless connection--some challenges in certain situations, but that's what you are doing all day long. Moment-to-moment challenges not that exhilarating. For some people that's probably very satisfying--homeowner worried it's not going to go well; at the end you see a beaming face, appreciative; solve people's problems, very satisfying. Might be some people have a personality where they enjoy doing something like that. Could be some people want variety.
14:36Government: while knowledge in economy is getting dispersed; as a result, power in the market place is being dispersed. Power in central government becoming less dispersed. What's going on there? Several things. Number of government units stays fixed: one Congress, 50 states. Meanwhile, the population within these units increases, increases scale. Amount of money per government unit going up--each state is spending more per constituent; Federal government spending more per constituent. Dollars per constituent going up, multiply that by constituents per legislator or component going up, get an explosion in the overall scale of government. Scale creep. In addition, get scope creep: government seeming to become involved in things at least in the United States it didn't used to be involved in. Didn't used to be an Environmental Protection function, didn't used to be as involved in providing health insurance. At the Federal level, didn't used to be involved in education; becoming more involved. Wouldn't one of the relevant measures be the size of government employment? You point out in the book the scope of Senators or Congress representatives to spend money has gotten bigger; only 435 members of Congress; that hasn't grown. Only 100 Senators, that hasn't changed. Could be they are more efficient at allocating. Think of employees that disperse the money. Size of government at the Federal level hasn't grown that much; state and local have grown tremendously. Tension: you'd think state and local government would be good that they grow--closer to the constituents. Not obvious that they are; probably run less efficiently; less oversight, less monitoring. When I first started to write this book was going to use as hook the county where we live, Montgomery County, MD, where we have 9 county council members allocating a budget of $4.5 billion. Very few people who could spend $500 million apiece a year consistently. Nobody's heard of these people, not even household names; could maybe name two of them or three. Incredible amount of power, very little checks and balances. County run by teacher's union--impression. Teacher's union follows it closely. People running for office in my area have yard signs with a little apple on it saying "Teacher Recommended" as if that's a selling point. Special interest. Good for teachers; not necessarily good for education. Not necessarily even good for teachers. Apple with a line through it as symbol for candidate. Incredible amount of power; all one party. Little checks and balances. State and local government: problem, checks and balances can often work poorly. Governmental units are not what they used to be. Can't call it "local government." Close to a million people in the county trying to elect these 9 representatives. That's why special interest group has so much power. The representatives don't know or care about their constituents as individuals; only care about this large aggregate. Montgomery County is about the size of one of the largest cantons in Switzerland. So if Montgomery County were in Switzerland, the legislature would have 150 people in it and it would be a Canton--there would be a local government underneath that. Would be like a state. You are suggesting that we need a larger government--measured by legislators or representatives to disperse their power somewhat. That's one direction. Or less spending would also do it. From a democratic point of view, to recreate the world of 1800, you would have to divide this country into about 250 equal-size state and each state would have to be divided into smaller units. Or if you wanted this country to look like Switzerland, last example of a truly Federalist country in the world, would have to take a state like Maryland and divide it up so each county would be sort of like a state with a legislature of 100 or more people and each county would have to further subdivide. Sounds terribly inefficient. Could be that scale in government is not a bug--it's a feature. From standpoint of democratic equality, almost clearly a bug. If you have 9 people running a county of a million people, that's an incredible concentration of power. Whether that's efficient or not, hard to feel you are equal to those 9 people--that they are the same, your neighbors. Special interest groups: taxi cab business, a lot of donations that come from one taxicab company, and they have a monopoly in the county. Bizarre idea. County makes it extremely difficult for a large store to open--have to have a special permit; no super-Walmart. Groceries not well provided. Real estate, graft thing. Cynical. Developers understand which political party their bread is buttered on.
24:54Potentially this high scale could be a feature. Efficient. School districts example. People have studied the scale of school districts. Much smaller than Montgomery County. Yes, it's inefficient to have a school district to have a school district with 150 students in it; but you don't need 100,000 students to have an efficient scale. Maybe the efficient scale is in the 10,000-20,000 range. In this country, we've consolidated school districts to the point where many are larger than that maximum efficiency. All you gain is an increase in power. Tradeoff in responsiveness to the will of the electorate. We see this in economics all the time. What's the efficient size of a restaurant? Not one that serves 4000 people. We see franchises that brand; emerges from the marketplace. Not clear for a school system that it emerges from anything other than the pursuit of power.
26:28Challenge viewpoint and raise alternative view. In the face of all this specialization going on in the private sector--two trends, specialization in private sector and concentration of power and influence in government sector--a lot of people argue exactly the opposite. True there's a dispersion of knowledge, but there's an increase in the centralization of power in the private economic sector. So, for example, there's economies of scale in search engines, media world. Lots of specialization, but the umbrella for the specialization, Google or Disney, people worry it offsets the dispersion you are talking about. Got to have this centralization of power on the government side to fight the centralization of power on the economic side. Start with last one. Don't think much evidence that having a strong central government weakens concentration of economic power. Actually leads to the opposite. In actuality, powerful government and powerful private sector are friends. Classic example: financial crisis, Goldman Sachs, Freddie Mac, Fannie Mae--government will do anything but put them out of business or even allow them to go out of business, so even when they fail the government won't allow them to go out of business. To say that big government defends the little people against them is backwards. Economies of scale: there really are economies of scale that utilities, communications companies, maybe Google can take advantage of. Nothing wrong with economies of scale emerging. Sometimes they go away; people find ways to overcome them. Somebody who seems to be on top, like IBM, seem to be fantastic scale economies in the computer industry 30 years ago--thought we had to break up IBM, but it turned out we didn't have to worry about breaking up IBM. Think that process emerges okay. In the case of government, the scale economy quite artificial: government says: We're going to do this--have large school systems--and it's not because the large school systems have outcompeted small systems, but we'll do it just because we can. Similar example, EconLog and in book: need a systemic, global regulator. Has a certain appeal: need more generalist in that dimension. In something like the financial system, the expertise you want to have to regulate the financial system just plain does not exist. When the subprime mortgage crisis hit, Bernanke and others thought that it would be a very confined crisis, hit housing. Many people thought that. Turned out that a lot of people regulating the system didn't understand what I would call the "embedded put options" that companies were writing to a great extent. AIG being the classic example. Even in September and October of 2008, watching Bernanke and Paulson, wincing at some of the things they were saying; understand enough of mortgage finance to know that some of the things they were saying were completely wrong in understanding how these instruments work. You are presuming they were trying to understand it. Bizarre situation: making public pronouncements that are supposed to calm markets, or maybe disguise what they are actually doing, right? Think they genuinely in this didn't understand. Don't think Paulson was cynical when he said "We are going to buy up the toxic assets." It was only after they passed the Troubled Asset Relief Bill (TARP bill) that enough experts were able to convince him that these are not the kind of assets you can just buy up, like bonds. Real demonstration of a lack of expertise by those who had the power. Didn't he check before? Not an indictment of Bernanke and Paulson that they didn't know as well as you did, who worked at Freddie Mac. The indictment is: Why didn't they draw on the knowledge of others in advance? Next question is: Why didn't they have to resign, why not asked to resign, resignation not tendered for that issue? Separate issue. Want to have this super-regulator who will know who to ask, ask the right people, obtain the right information. Have the good database. Just illusive. For the reason Hayek explained during the socialist calculation debate: you just cannot put all of the information into a central computer and outthink the dispersed information available in a market. Frustrating for people to think about that. They prefer it not to be true. At end of this month, invited to participate in a conference on the American labor force; get sense, maybe will be wrong, but will feel like the token Negro at a Klu Klux Klan rally, because my guess is everyone there is going to have what they think of as a solution that the government can implement to make it work centrally. I'm going to ask, what if you didn't have the government to back your solution? Could you raise money as an entrepreneur and implement this solution? Well, I could but people won't pay for it. Well, could you convince me as a charitable donor to invest in your solution? Idea that a decentralized, trial and error approach to solving a problem is just very counter-intuitive to people. There's an intelligent designer out there who can solve it centrally. An expert.
35:03Push back on that idea in a different way. Article in The New Yorker by Atul Gawande: talks about this, podcast on health care. Senior citizen comes to office with a bunch of problems, gets sent to a bunch of specialists; each talk about thing to fix, each prescribe drugs, therapies; but sees a generalist, who sees some of these solutions won't work with others. Rather than looking for a nail to use with the hammer, prescribes a more general set of solutions that work for her. Gawande's point is: need more generalists, especially for old people. Current health care system destroys incentives for that kind of solution; can understand appeal of mandating it as a solution. Could you argue that in the regulatory or voluntary world, we need to subsidize generalists? True they can't master all the details, but don't we need people to oversee these complex systems? No individual has the incentive to care about the fragility and stability of the whole system. Isn't that a potential role for regulators? Horrible argument, but hear it a lot and thought provoking. If you find this almost-omniscient generalist, let me know, because in theory it's a great idea. In practice what you get is either people whose view of themselves or whose expectations people place on them are unrealistically high. Usually combination of both. Somebody who had a realistic understanding of their limitations probably would not find their way into a position where they are expected to do something they don't think is possible. Trivial example: estimating the multiplier from the fiscal stimulus. If somebody said, "Arnold, go on CNN and tell them what the multiplier is for the fiscal stimulus," I would babble. I'd have no idea. It could be negative, could be positive-4; could be all over the place; don't think it's an estimable number. On the other hand, Mark Zandi comes on there and says it's 1.5. Well, he's the one who makes it on TV. Those are the types of people who make it into positions of power. So if you believe in this omniscient regulator, you get what you pay for--get somebody who thinks they know more than they really do, and who lives with that expectation. Paulson: people expect me to have all this knowledge, act as if I have all this knowledge; that's how I've gotten ahead so I'm going to go ahead and do it. Will probably never admit, even afterward, that he lacked the correct knowledge. Memoir. Like Nietzsche's chapter: Why I'm so smart, why I'm so strong. React to that. Couldn't you argue that a CEO is a generalist? In the old days, say an energy company, CEO a former engineer, sometimes a chemist. Specialist, suddenly find themselves in a generalist position; know a lot about the core of business, but inevitably little about who set of stuff--personnel, finance, IT. Flawed, one human being, but in competition with other CEOs in other companies; we rely on that competition to make that work pretty well. Maybe that's a model we ought to move toward--try to introduce more competition in government. Think of the checks and balances in place in the market when a CEO can do more than he really can. Buys a company in an industry he doesn't understand. Chances are he's going to mismanage it; share holders are going to punish him for doing that, or try to snap up that mismanaged company at a lower price and manage it better. Fight for corporate control, tend to limit, provide a check and balance. Imagine a world with no check and balance. CEO says I'm capable of doing x: take this energy industry and diversify it into finance and just does it with no checks. Where would you end up? Probably with a lot of inefficient, large companies. That's my model of government. People just say by fiat: we are capable of doing this, regulating drugs, urban planning, grid full of electric cars. Without any check, without anyone being able to say no.
43:48That isn't really the way we regulate anything. We have a weird set of checks and balances, not well-designed. Examples in regulatory world: Most regulations are not very specific. Specificity is turned over to the agency--not an expert, not the head of the FDA who writes the regulation, not Barney Frank, not President Obama in the case of health care, but a whole bunch of individuals who have expertise in the areas they are involved in. They write the details and then it goes through bizarre process of adjudication and people fight. Process of back and forth. If they mess up, push back from members of Congress who are their constituents. Byzantine, specialized, but exists. Typically who wins in that process is the firm with the biggest lawyers. Recent crisis: Bernanke and Paulson did many things without the checks and balances. To some extent the people acquiesced, though quite a bit of anger about it. This election is the first chance people have to vote against the bailouts. To some extent overreaction to the change, but it's the speed at which it occurred. Financial industry, auto industry. Not on the road to socialism, but first step; 100 yards down the road. Certainly an increase in the scope of government, doing things it didn't do before.
47:05Ways that government might exploit some of the improvements in knowledge and information. Role of the internet in changing access to information. What could we do to take advantage of that? Internet increases ability of specialized knowledge to assert itself. Nothing stopping individual from putting up their website or application. Easy to aggregate information in different ways: Wikipedia to unknown niche sites. Done in a very decentralized way. Contrast with mass media of the 1950s and 1960s, radio and television. If you wanted to communicate you had to pay a huge amount of money to pay a license, get broadcasting equipment, radio tower. Only a few people could send out information. Now anybody can. Creates advantages for decentralized knowledge. Fewer advantages for concentration of power. The President can't give a Fireside Chat and have everyone listen to it and nobody challenge it. When the President speaks, fewer tune in; of those who tune in, a lot are critics and instantly post about it on the Internet. They get their say. Different environment. Internet is an interesting model of an almost-anarchy situation. The only central control is domain registration, Internet Corporation for Assigned Names and Numbers (ICANN). If you want to get or you have to go through this domain registry. If you join this system and link up according to the protocols, you are on. When people need to solve generic problems on the internet, there are Internet engineering task forces, at least 10 years ago: assemble to solve a problem, say to standardize Javascript--maybe 15 years ago. Set up a task force to standardize Javascript. Volunteers, kind of set itself up. Group of people gets together, people with strong interest in it, meet, negotiation, come up with proposed solution. When people stop complaining about that solution, it stabilizes. Can you imagine government working like that? If we could have that work in actual government how much better it would be. Private sector example: more nuance into the specialization thing. Want to use Skype, have a Mac, download it, go to connect; it doesn't connect. Throw it out, try again. Nothing works. One way to solve that problem in 1950 would be to call the Skype repair person; he would come to the house. I don't do that. Get on the web; Google: skype mac os can't get in. Pull up a bunch of pages from 2007, people had a problem. Some on websites: Answer my problem. Try it. On about the 7th page of the Google search, find somebody who says you need to throw out that folder. Go to son, says shouldn't throw it out. Went further down Google page; someone says he tried that and then realized he needed to throw out an even bigger subset of folders. Tried that; it worked. Fixed problem myself. I'm not a specialist. Becoming a generalist. Internet allows that problem solving. Clay Shirky podcast: people who used to spend their time watching television can now spend their time creating public goods like answers to problems like how to fix Skype. Also helped me fix how to unclog my bathtub without having to call a plumber. Took a while, maybe half an hour of Googling around, a lot of solutions didn't help.
54:37Back to government. Talk about a task force that could disband. Could we imagine some problems government currently solves by command and control that could be solved this way. Imperfect knowledge. Some people say the financial crisis caused by deregulation. You pointed out that then the answer is clearly: just reregulate. Yet that's not what we did. Enormous disconnect, political reasons. For the record: I didn't say that the solution is to reregulate; I said that if you think the problem was we deregulated, then the solution is to reregulate. Find those regulations and stick them back on. Examples where there would be a solution to the problem that would be more decentralized? Suppose government got out of the meat inspection business. Egg inspection, salmonella. People are aware that buying eggs at random could be dangerous. Egg manufacturers would have a really strong interest in supplying certified eggs. Would look to somebody to create an egg-inspection service. Would have to establish it is credible. Like Consumer Reports. Might even have more incentive than government. If a private agency had messed up like this, they'd be out of business. The government--don't know if anyone got fired over this. No systemic check on that. Ratings agencies, which many people say failed in the financial crisis, are still in existence. Reason for that? Why didn't Moody's, Fitch go out of business? Part of the reason is that where they messed up was in rating securities that were not going to be traded in the market place. Purpose of the ratings was to get regulatory lower capital requirements. Customer was the regulator. When Moody's rates a bond, they rate it on behalf of the corporation selling the bond, but there's a customer out there, in some sense the buyer of the bond. If Moody's betrays that customer, at some point the customer will stop using those ratings. General Motors (GM) stops using Moody's to get bonds rated. But when the customer was the regulator the customers didn't care. At some point somebody introduced something, an amendment to the financial reform bill, saying that regulators could not use ratings. Excited, thought it even passed at one point; but when finally negotiated between House and Senate, not there. If the rating agencies were to mess up in rating bonds that are actually traded as badly as they messed up in rating bonds to get past regulators, and not supported by government creating this oligopoly, I think they would go out of business. Think GM is required by law when they issue a bond to be rated; that is what props up these folks. That and the license to be a rating agency is issued by the SEC. Small group of agencies. Like the accounting business, big four accounting firms. So much pressure on big companies to have proper accounting. Sense that if one more goes out of business we won't have accounting; so they are protected from going out of business in same way as ratings agencies. In the case of the ratings agencies, something happened two months ago that was almost ignored. In the new financial reform--let's call it the new set of financial regulations--there were penalties assessed on the ratings agencies. Not sure how that's actually enforced. Can give something a triple-A rating and it goes bankrupt. Can mis-assess. But if you systematically mis-assess, new regulation, stark penalty, don't remember what it is. Fines, for "bad ratings." They immediately said, when Ford was going to issue a bond, that we're not going to rate this, because it's too scary. Bad time. Government waived the requirement that the Ford bond have a rating! Of course people bought it without the rating. Illustrated that this is a weird, not normal competitive market.
1:02:30Lot of creative ideas in book, use of vouchers. Most ideas would be considered unviable politically. Could push, maybe over next 10-20 years, maybe seen as more viable. Other is to be more activist. Term limits. Any ideas what might work? Ways to go in terms of changing culture and mindset of both elites and masses. For masses, ordinary people, case of opening eyes to possibilities. Virtual government: could sign up and say I belong to this government for the purpose of welfare programs and education. Could be that this government is run out of California even if you live in New Jersey. Garbage collection could still be public, but why have it be only related to your physical jurisdiction. Look at lawn care. We manage to get an emergent order with different lawn care people operating. Getting people to think in those terms. For the elites, gradually losing some of the illusions they have. Example, 1970s, when the elites thought that wage and price controls would be a solution to inflation and unemployment. Throughout the 1960s they kept saying try it, try it, try it. Tried it in the 1970s, results were horrible. No one advocates wage and price controls as a solution today. Ability of the human mind to tune out negative feedback is tremendous; confirmation bias is great.

More EconTalk Episodes