« Remembering Mike Denny | Main | When Will Low Interest Rates End? »


Feed You can follow this conversation by subscribing to the comment feed for this post.

"But the government takes no action to ban the promotion of those financial assets in future. All it does is impose regulations that reduce, but do not eliminate, the chances that the promoters will be unable to keep their promises.

"That is even harder to understand."

Can you spell P-L-U-T-O-C-R-A-C-Y, boys and girls?

"Meanwhile, the government itself produces a financial asset that people can buy. And the government promises that the people who buy that financial asset will never lose money by doing so. And the government can keep its promise. Because that financial asset is money. It is the perfect asset for people who suffer from nominal-loss-aversion.

"But instead of promoting its own financial asset, the government helps the promoters of competing financial assets to keep their false promises, even when it is costly for the country for the government to do this.

"That is incredibly hard to understand."

Can you spell P-L-U-T-O-C-R-A-C-Y, boys and girls?

People have an ideological problem with the government issuing debt. It is their preference to accomplish the same thing in a less robust way through private actors. It is my preference to avoid the problems they inevitably create, but that is impossible because markets are interdependent.

I agree with Min. To flesh out his point…..

Printing money and lending it out is profitable. That’s what private banks do. But it’s a risky activity, as Nick points out. So to ensure that that activity is curtailed as little as possible, banks devote HUGE AMOUNTS of money and effort to brainwashing and hoodwinking politicians. E.g. the UK’s finance industry spends £90million a year on lobbying politicians. See:


As to the US, Senator Dick Durbin once said “And the banks . . . . are still the most powerful lobby on Capitol Hill. And they frankly own the place.”

Or to employ Min’s style, can you spell R-E-V-O-L-V-I–N-G D-O-O-R boys and girls?

"But that problem is easily solved, if the promoter simply invested all the proceeds from selling his financial asset into the government's financial asset"
The policy has shifted to this direction with derivative regulations:

Nick wonders why government doesn’t promote it’s “own financial asset”. Part of the explanation is set out by George Selgin in this article:


Selgin argues that if monetary base was the only money (the government’s “financial asset”), private banks setting up in that scenario would print and lend out sufficiently large amounts of their own financial asset that the value of base money was inflated away to near nothing: to the level where private banks had just enough base money to enable them to settle up between themselves. I'm not sure that Selgin is 100% right there, but I think he's onto something.

That process is also alluded to by Joseph Huber and James Robertson here:


See paragraph starting “Allowing banks to create new money..” (p.31)


"Instead, whole countries get into very bad trouble because of people who suffer from nominal-loss-aversion, the promoters who make promises to those people that everybody knows they cannot always keep, and the governments who enable them to do this and keep those promises for them."

"It must be some sort of really weird co-dependency thing."

Kind of like raising children - eh? Child makes a promise that he has trouble keeping - parent is there to ensure that the promise is kept.

Ralph: fractional reserve banking presumably reduces the demand for central money. And if the central bank kept its stock of money constant, that would increase the price level. But any sensible central bank would simply reduce the stock of its own money in response, to prevent the price level rising.

Plutocracy hardly explains regulation to protect incumbent workers (all those restricting-dismissal laws), taxi licenses or to ration land use for housing (the Zoned Zone, in Krugman's felicitous phrase).

Regulation has a strong tendency to protect incumbents (see aforementioned cases)--whether incumbent labour providers or asset holders--and politicians compete for votes and funding to purchase access to voters while people in industries regulated with high levels of official discretion (housing, construction, finance) or regulation-based property rights (Disney) need to purchase access to officials. No more complicated model or ideological baggage needed.

Also, treating economic exchanges as if only the "real" story counts is acting as if money isn't a medium of account. For things such as debts, for example. Money is the language of transaction, if you will, and do we think that language doesn't count, that it is only the "real" referents underneath that matter?

Perhaps another way to look at it is that money lowers transaction costs and "money illusion" is a cost of money lowering transaction costs.

I would add to Lorenzo's point re: plutocracy that £90 million of lobbying money doesn't indicate to me that people are the rulers, but that they are the supplicants. Kings don't give tribute to their subjects; subjects give tributes to their kings; the influence of being rich is because larger tributes get more royal favour. And this is unfortunately an unavoidable aspect of human society, because all societies have structures of power, and whoever can offer more (wealth, influence, soldiers etc.) will have more sway with the powerful. The key problem of politics is moderating these realities.

That's why the contemporary financial system, being a very dodgy half-way house between the market and nationalisation, is appalling but not surprising.

Fractional reserve banking is better:

One of the problems with 100% banking is that the central bank balance sheet size would be enormous. Where to invest those central bank assets? It is better to outsource these decisions to fractional reserve banks.

A related problem is the bank equity problem. With fractional reserve banking, market allocates capital to the commercial banking system, while the government allocates capital to the central bank. 100% banks do not need capital, and the only place where capital is invested with 100% banking is the central bank. Do we trust Treasury to make right decisions here? We would get either too much capital in the central banking (as in Japan before Abe), or too little (current Eurozone where ECB has inadequate balance sheet far away from what Friedman rule would recommend).

"I own bank shares". Better than airline shares! Around 2003, the *cumulative* profits of the entire US airline industry was 0.

You should prefer a certain return to a variable return even if you do not have a diminishing marginal utility of consumption. Even if they have the same expected arithmetic return, certain returns offer better long term geometric returns, and geometric returns are what you take home.

Incumbency is just short form plutocracy, which is why doctors get much more preferred treatment than union workers.

Lord: than public sector union workers? After one takes out scarcity value?


Why do think doctors are scarce?


I don't think people suffer from nominal loss aversion. I think people suffer from Cassandra nominal loss aversion - they fear suffering nominal losses without having the power to recoup those losses. Tell a gambler that he has a 50 / 50 shot a winning a particular game and he is fine with it. Tell a gambler that he has a 50 / 50 shot at winning the next game but he must never gamble afterwords and he is not fine with it.

Lord: because the % of the workforce capable of being doctors is somewhat smaller than the % of the work capable of being public sector workers, and even more so with surgeons.

Any further scarcity of doctors probably has more to do with the cost of university education and liability insurance.

If you are trying to saying richer and more articulate special interests are going to be somewhat advantaged; well yes. But describing that as "plutocracy" rather misses out the broader pattern of favoured interests. For example, young and migrant workers generally have better opportunities in the US than the EU, (and in the UK than continental Europe) but would one therefore say the EU is more plutocratic?

To put it another way, complaining about "plutocracy" seems a way of noticing some favoured interests but not others.

In case it is not obvious, the last paragraph of my 5:51pm comment is actually directed to the post. :)

Lord: doctors are scarcer because the % of the population capable of being doctors (and even more surgeons) is smaller than the % of the population capable of being public sector workers. Plus the cost of university education and liability insurance.

The opportunities of young and migrant workers are generally better in the US than the EU: would we then describe the EU as more plutocratic?

To put it another way, complaining about "plutocracy" seems a way of noticing some favoured interests but not others.


You say, “any sensible central bank would simply reduce the stock of its own money in response, to prevent the price level rising.” Yes: that’s an alternative way the above Selgin theory could work out. In fact we might easily see that happening in the next few years: i.e. if private sector irrational exuberance returns, central banks will withdraw the base money they’ve distributed via QE.


You claim that under 100% reserve, the central bank has to find an “enormous” volume of assets to invest in. Why? Under full reserve (and to a lesser extend under fractional reserve) the government / central bank machine simply creates money out of thin air and spends it into the economy. As long as it doesn’t create and spend too much, inflation won’t become excessive.

You could argue that the CB / government then has a pile of liabilities but no assets. However the extent to which £20 notes are a liability of the Bank of England is very debatable. The notes actually say “I promise the pay the bearer the sum of £20”. But that’s meaningless. Plus government can simply expropriate money from the private sector whenever it wants via taxation. That’s not a normal debtor / creditor relationship.

Next, you criticise 100% reserve on the grounds that “We would get either too much capital in the central banking . . .or too little.” I don’t see what that means, since a central bank can be an entity with no assets and no liabilities (despite issuing millions of $100 bills). But if you’re suggesting that CBs might issue too much money with excess inflation being the result, then private banks’ record in that connection is even worse: we’ve just had a crises sparked off by irresponsible private bank behaviour. And in more general terms, why do we have CBs and governments regulate demand? It’s because we think they can improve on the boom bust cycle that the free market gives us.

Actually I do not think that it is all that hard to understand. You only require that people in general will be prone to some specific types of cognitive bias, for instance like this one.

So if people ignore low probability of bad outcome then they prefer interest bearing bank money to government money even if they are averse to nominal loss. So it just takes some bias that makes them blind to it.

But here is the thing - I do not think it is a bad outcome to do this kind of "paternalistic nudge" in this direction by government. The whole point of banking sector may be to induce the society as a whole to undertake more risky investments than it would be the case without banking.


let's take a simplified model and say that there are three problems in the central banking - (1) choosing the trend rate of inflation and accordingly trend rate of monetary base growth, (2) ensuring macroeconomic stability by deviating from the trend rate of monetary base growth (i.e. doing OMOs and reversing them to stabilize the economy), (3) choosing the right average ratio of money / GDP (Friedman rule is about this question)

I agree that 100% reserve banking system can get the question (1) right.

If the central bank creates money out of thin air, we have problems with (2) - how do we do reverse OMOs when needed? So it has to choose the best assets to invest in - this is a big problem in 100% reserve banking, as this decision is not outsourced to the fractional reserve banks.

We also have problems with (3) in 100% banking. Markets allocate capital to fractional reserve banks and help us to get the ratio of broad money/GDP right. With 100% banking, the only real bank is the central bank, and we would get either too much capital in the central bank, or too little (i.e. money/GDP ratio that is too high, or the ratio that is too small).


Re your No.2, I agree that “reversal” is a problem. However, I don’t agree that the central bank / government machine only buys assets with the money it creates. In fact the bulk would be spent to standard public spending items: roads, education, law enforcement, etc. So… how can that be reversed?

Well exactly the same problem applies to traditional fiscal stimulus: e.g. if fiscal stimulus takes the form of raising the state pensions, then pensioners might be up in arms if that pension rise is withdrawn. However, various forms of fiscal stimulus are easily reversed: e.g. the UK cut and then raised its sales tax (VAT) since the crises, and there were no riots.

Also, there is an element of automatic reversal built into the system as follows. Inflation is constantly eating away at the real value of the monetary base and government debt. Thus if government does a bout of stimulus, and then stops, that stimulus is automatically reversed in due course.

Re your point about central bank capital being too large or too small, I answered that point as best I could in an earlier comment. But just to repeat one point, central banks are very different to commercial banks, and in particular, I don’t think the word “capital” in relation to central banks actually means anything (although central banks obvious publish balance sheets in accordance with the rules of double entry book-keeping).

Or as Warren Mosler put it, central banks are like umpires in tennis: they produce points from nowhere and award them to players / households / firms.

central banks need capital to absorb risk if they do stabilization.

A large modern bank supplies a full range of financial services. Banks can get into trouble through abuses of any one of these. It can:

Make crony loans like the S&Ls did

Speculate in financial products like derivatives and options

Take on too much leverage, often by hiding the fact in off balance sheet entities

Exploit the conflict of interest between being a broker and provider of financial advice and selling those investments on which the bank made the most profit.

Have weak internal controls, which allow employees to lose enormous sums in huge speculations.

Assume too much tail risk like AIG

Engage in fraudulent securitization

Incur huge costs by engaging in illegal practices like laundering money for drug dealers or fraudulent foreclosure practices and improperly documented real estate operations or making material misrepresentations in securities documents.

Make huge investments in the bonds of their home countries where the nominal 0 risk was substantially different from the risk implied by the much higher interest rates the bonds paid. They borrowed money at low rates from the ECB to do this.

Suffer from the general problem of a country and its banks being joined at the hip.

Make zombie loans to keep from having to rate loans as nonperforming, which would hurt their balance sheets (as in Japan)

The problem, as I see it, is allowing banks to become enormous financial conglomerates. Very few of the problems have come from traditional FDIC insured fractional reserve banking. Most have come from all the other things we've let banks do.

Without the FDIC, the banks would probably create their own clearing house to provide similar guarantees. This system worked well for banks for 50 years and is used by many other financial products.

What you would lose are the regulatory functions (like capital ratios) and perhaps orderly liquidation (bankruptcy and checkable deposits are antithetical).

"Instead, whole countries get into very bad trouble because of people who suffer from nominal-loss-aversion, the promoters who make promises to those people that everybody knows they cannot always keep, and the governments who enable them to do this and keep those promises for them."

Gary Gorton IIR argues that this is the purpose of modern banks - to reconcile the low risk tolerance of investors with the higher risks inherent to financing a modern economy through obfuscation. That is - to hide from customers the risks they are assuming.

Unfortunately, this gives rise to the evils I listed above, which can be generally described as taking advantage of a privileged position to exact rents by any and all practical means.

Nick, I think the issue is this. Commercial banks provide two functions:

1. deposit / payment system
2. short-term commercial credit (inventory financing etc)

If you insist on 100% reserves, then what happens with function 2? I assume the economy still needs it, so let's split commercial banks into two pieces: "deposit banks" that are deposit-funded with a 100% reserve requirement, and "credit banks" providing commercial credit, funded through debt securities. Clearly, credit bank debt securities could be purchased by the Fed in OMO, so Vaidas' point is completely invalid, i.e., there is no problem with outsourcing credit decisions to credit banks. The Fed can expand by buying securities of these banks on the open market and contract by selling them.

I think this is workable, but what's the benefit? Yes, the deposit banks (with the 100% requirement) are now safer than today's commercial banks are, but the commercial credit function is also essential to the economy. If a financial shock disrupts the debt-funded credit banks and commercial credit seizes up, the Fed will have little choice but to "bail them out" (provide TARP-style massive liquidity). So what have we gained relative to today? If we have to bail out credit banks regardless, why bother separating out the deposit function? What's the harm in leaving credit and deposit functions conjoined in commercial banks?

It may be that fractional reserve banking is an artifact of the gold standard, where fractional reserve banking's ability to influence the money supply was extremely important, whereas today is unneeded because the Fed can manipulate the monetary base directly. But it seems like a harmless artifact to me, i.e. zero benefit from "fixing" it.


Kenneth Duda
Menlo Park, CA

Kenneth, the benefit is that you can get rid of deposit insurance (and the arbitrary and political pricing of the insurance).

I agree that it wouldn't do much, if anything, to prevent a financial crisis.

It all has to do with history.

If you have a government with a large national debt already, then it could be funded with interest bearing reserve balances.

By paying sufficiently high interest, it would seem possible for the government to crowd out fractional reserve banks.

Why isn't this done?

Perhaps it is because governments got into this business in order to borrow at below market interest rates--zero nominal rates and more or less zero real rates to start. Now it is negative real rates.

Compelling banks to hold reserves means the government can borrow more at the negative real interest rate.

Banks and their depositors have every incentive to reduce reserve requirements so that they don't have to lend as much to the government at zero or negative interest rates.

And all of that comes from a history of a gold standard where the more fundamental reserve bears no interest and government is requiring gold reserves so that it can borrow them during wartime to fund imports.

If the government already has a sufficiently large national debt, so that the demand to hold bank reserves at an interest rate high enough to crowd out fractional reserve banks is less than the national debt, then there is still other sorts of government bonds to allow for open market operations.

Or, the interest rate on reserves could just be varied to keep inflation or nominal GDP on target.

Are there any problems with this? Well, the government is borrowing very short. If you assume the government has perfect credit, it isn't a problem, but that isn't necessarily correct.

If the government doesn't have perfect credit, you don't have to continually refinance as much of the national debt. The huge run up in the interest rate on reserves needed to get people to hold them would be costly to the government. The tax hikes or government spending cuts needed immediately would be greater than in a situation where much of the debt is long term and only a bit coming due in the near future.

Personally, I don't favor a system where the government debt needs to be high enough to meet the demand to hold money.

I favor full privatization of money.

Unless the national debt is very small, there is no problem with 100% reserve banking. The central bank just buys government bonds, and issues more currency (and reserves, same thing). This would become a problem if the central bank ran out of government bonds to buy. Then you would either have to run a deficit, to increase the national debt, or the central bank would have to get into the business of buying other assets. The people who previously held the bonds can either hold currency, or deposits in 100% reserve banks, or start making loans, or buy shares in financial intermediaries that make loans.

Financial crises don't matter (much) unless they cause widespread defaults that affect the monetary system. The 1987 stock market crash did not cause a recession.

As in all cases of government intervention to make the market work better, we need to ask the question: where's the externality?

Bill: "I favor full privatization of money."

Trouble is: what Buchanan called "the Samaritan's Dilemma". If the result of privatisation is lots stupid people believing stupid promises, the government will come in to rescue them when those promises go bad. The government's commitment not to do so is not credible. To avoid Samaritan's Dilemma, the government might need to ban those stupid behaviours where it knows its own commitment not to rescue people from the consequences of their own actions will not be credible.

Milton Friedman advocated 100% reserve in 1960. I’ve put a relevant passage from his book online (about 300 words):



You say “central banks need capital to absorb risk if they do stabilization”. CB’s capital (if you want to call it that) consists of the ability to print any amount of money anytime. And if we consider CBs and governments as a single entity, they can grab any amount of money anytime off the private sector.

Kenneth Duda,

You argue that what you call “credit banks” might need rescuing in the same way as our existing banks need rescuing, thus the move to 100% reserve is not worthwhile. That question was answered by Mervyn King is his “Bagehot to Basel” speech As he put it:

“And we saw in 1987 and again in the early 2000s, that a sharp fall in equity values did not cause the same damage as did the banking crisis. Equity markets provide a natural safety valve, and when they suffer sharp falls, economic policy can respond. But when the banking system failed in September 2008, not even massive injections of both liquidity and capital by the state could prevent a devastating collapse of confidence and
output around the world.”

I.e. all that happens when your “credit banks” do badly is that their share price falls. That’s not nearly as serious as a potential collapse of the banking system. Plus if problems in the “credit bank” industry do spark off a bit of a recession, I don’t see why governments should assist those banks rather than provide general stimulus, any more than a collapse in car sales is a reason for government to subsidise car makers. If credit banks go thru a bad patch, it could easily be they’ve over-sold their product, i.e. lent too much. Indeed, that was pretty much what sparked off the recent crisis, wasn’t it?

Ralph: "CB’s capital (if you want to call it that) consists of the ability to print any amount of money anytime. And if we consider CBs and governments as a single entity, they can grab any amount of money anytime off the private sector."

No, CB's capital consists of the ability to print any amount of money anytime minus the propensity of the treasury to divert this money for other purposes. So my point is that treasury rarely gets this difference right. Either too much capital is invested in the central bank (like in Japan before Abe), or too little (like in Latin America). Capital inflows to the fractional reserve banks (& outflows) mitigate this.

"Financial crises don't matter (much) unless they cause widespread defaults that affect the monetary system. The 1987 stock market crash did not cause a recession."

But widespread defaults, bank runs and shortage of commercial credit are what make it a crisis. 2008 was a crisis. 1907 was a crisis. 1987 wasn't.

" the promoters who make promises to those people that everybody knows they cannot always keep, and the governments who enable them to do this and keep those promises for them.

It must be some sort of really weird co-dependency thing."

"Everybody" doesn't know. What makes it work is that the promises are believable, though, of course, the less sophisticated the audience, the further belief can be stretched. People have tremendous faith in "experts" and no idea how to recognize one. Look at the faith people had in Madoff. These were sophisticated investors by anybody's standard. You'd think accountants running hedge funds would know better. And how about his banks? He clearly wasn't trading enough to be legitimate. It wasn't even close. A bank knows what a hedge fund account should look like

It's like the disappearing ball trick. The magician throws the ball up in the air and catches it a few times then it suddenly disappears. The last time he never threw it, of course. People became accustomed to something and saw what they expected to see. Madoff's consistently good results were seen as proof of his investing genius, not as too good to be true.

After all, we take the enormous and rapid growth of the financial system as proof of its important economic contributions, not as a sign that it must be rent collection by a bunch of crooks. This despite what you would think was enough legal evidence from court records to convince anyone. Fraudulently notarizing documents. laundering money for drug cartels? Well nobody is perfect.


This is an interesting post - with an awfully weak conclusion.

"It must be some sort of really weird co-dependency thing."

You're a very clear (economics) thinker/teacher - from my regular reading of this blog, which I really like - so why do you seemingly give up when it goes outside of the economic realm, e.g. into politics.

There are a number of politics models (ok, let's call it "political science") which easily explain these observations:
- the overall system-wide outcome may be suboptimal
- but many individual actors do very well out of this
- in particular, financial intermediaries
- and, in fact, all the agents throughout the system (advisors, brokers etc).
- but the people who make the rules are easily influenced, through a wide range of power devices (expertise, hiring, payments, hierarchy)
- You can look at Mancur Olsen-type interest group models, or more recent "quiet politics" models (i.e. when you're making regulations in esoteric areas, the rules tend to be made by the experts who work in the industry).

Interesting, worth digging into further.

Andrew: thanks. Yes, I wimped out on the conclusion. But I didn't have anything interesting to say, that anybody else couldn't say better, so I left it there.

That's ok - understandable! (Though I feel as if the entire economics profession has wimped out on this, hence my frustrated comments...).

Nick: Fed governor has just delivered an interesting speech where he defends banks against your charges: http://www.federalreserve.gov/newsevents/speech/stein20140103a.htm


The answer lies in what you have omitted. We have come to believe (or rather we have been told) that government is unable to allocate capital efficiently, so it must be done by the private sector. Therefore we need private sector intermediaries that provide safe financial assets to people who want to save, and reallocate that capital to people and businesses who want to borrow, we assume on the basis of some kind of sensible assessment of risk versus return.

The truth is that 1) those private sector intermediaries have proved themselves to be appallingly bad at assessing risk versus return sensibly and allocating capital efficiently 2) those private sector intermediaries are unable to guarantee the safety of the funds people place with them. In short, they do not do the two things we expect of them. Yet rather than saying "this is ridiculous, get rid of them and let's do it ourselves", governments illogically provide the suppliers of funds with guarantees of safety so that the intermediaries can continue to misprice risk and misallocate capital.

Is this a fair summary of your post? If so I sympathise. I have walked this path already.

I have three things to say:

1) We need to stop pretending that the private sector is "better" at assessing risk and allocating capital than the public sector. There is zero evidence for this. Yes, there have been some awful examples of public-sector banking - Spanish cajas spring to mind. But you can hardly say that private-sector banking has been a resounding success either.

2) We need to stop pretending that the public sector is constrained by arbitrary debt and/or seigniorage limits. That's what inflation targeting frees us from. We need to trust central banks to do their jobs.

3) And we need to stop pretending that the private sector can create safe assets. Only government can really create safe assets in sufficient quantity (cf Gary Gorton). Let it do it.

When I looked at the future of banking on Pieria about six months ago, I remember writing that I never, ever expected to find myself recommending state banking - but that's where I ended up. Logically, if the only way banks can operate is with the backing of a system of state guarantees, they are not really private sector institutions at all.

It seems to me that the people who insist that banks must remain private-sector institutions are those who at heart don't trust government. And that includes some of the people who ARE the government. Talk about cognitive dissonance.

I bet you never expected to hear this from me of all people!

Frances: and it also implies that remuneration in banking should be that of the public sector. Let's say $ 200K for the assistant-deputy minister in charge of deposit creation (formerly bank president. Those who do not agree are free to go exercise their entrepreneurial talents wherever they feel welcome.

And banking remuneration should be at public sector level. A deputy-minister for deposit creation ( formerly bank president) could get let's say $ 200K.

Frances C: Australian states turned out to be god-awful at running banks too.

Nick, does this post mean that you're sympathetic to the fringe movement known as the "Greenbackers", who seek to all ban fractional reserve banking and then have the Treasury directly issue money? There's an animated movie about it here: http://www.youtube.com/watch?v=0K5_JE_gOys
I think there have been extensive refutations of their views.

@Keshav: I watched the youtube - quite entertaining. As stated in the video: it is not a matter of tinkering with the fractional reserve system but rather a wholesale replacement. You might be interested in Milton Friedman's "Proposed Mandate by the Congress to the Federal Reserve System" [Monetary Policy and the Management of the Public Debt, Joint Economic Committee, 82nd Congress 1952], which Friedman describes as a summary of "A Monetary and Fiscal Framework for Economic Stability" American Econmic Review June 1948 pp 245-64. Friedman describes a system with 100% reserve plus an explicit role for government in controlling the supply of money. I wonder why he gave up on that idea?

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad