Bad Inflation Bets and Why They Were Bad

Brad Delong rightly slams Austrian economist Robert Murphy this morning for a bet he made in 2009 regarding inflation.  Murphy stated that headline inflation would hit 10% by January 2013.  Well, here we are with 24 hours to go and the latest monthly CPI reading is 1.8%.  I don’t want to just pile on Murphy with personal attacks.  Instead, I think it’s constructive to understand why this prediction was wrong because it’s at the heart of an important economics and finance understanding.

If we jump in the Google time machine we can see what was said back in 2009 that was so wrong.  Murphy was working from the same premise that many economists work from.  He saw the Fed flooding the banking system with reserves and assumed that this would cause inflation.  He said:

“In order to keep those reserves from working their way back into the hands of the general public (where they can start pushing up prices), the Fed will have to raise the interest rate it pays to persuade the banks to keep the reserves parked at the Fed. But this simply postpones the day of reckoning, as the troublesome stockpile of excess reserves grows even faster.”

This is not correct and it displays a huge flaw in the model that Murphy is working with.  It’s worth noting that Delong and others are working under a model that actually isn’t that different (though their “liquidity trap” theory has stated that the Murphy model is temporarily broken).  Both models are wrong.

Monetary Realism starts from an understanding of modern banking.  We understand that the US monetary system is essentially privatized.  In other words, the money supply is controlled almost entirely by private banks whose ability to create loans creates deposits which are the primary form of money we all use.  The money supply expands and contracts (mostly expands) in an elastic form based on the public’s demand for loans.

The flaw in the Murphy model is that he assumed that reserves are somehow related to a banks ability to loan money.  He specifically shows the scary chart of M1 going parabolic and then states in clear terms that this money will work its way into the public.

This is really important so I am going to cover this point again.  There are two types of money in our monetary system.  Banks deposits (the money we all use) are inside money because it is created inside the private sector (controlled by an oligopoly of private banks).  Outside money facilitates inside money and exists in the form of cash, coins and bank reserves.  This money comes from outside the private sector.  It is supplied by the government to facilitate the use of inside money.  Cash, for instance, is issued by the US Treasury to allows member Fed banks to stock vaults for customers who wish to draw down their bank accounts for transactional convenience.  Coins serve a similar purpose.  See here for more.

Reserves are a bit different.  Reserves exist solely because of the Federal Reserve System.  And they serve two purposes – 1. helping banks settle interbank payments; 2. helping banks meet reserve requirements.  Bank reserves are just deposits held on reserve at Fed banks.  You can think of reserves as existing in their own market that is totally separate and inaccessible to the non-bank private sector.  In other words, reserves are the money banks use to do business with one another.

But more importantly, banks don’t lend their reserves.  Banks lend based on their solvency or capital constraint.  Reserves are merely an asset of the bank.  When the Fed implements monetary policy like QE they don’t change the capital position of the banks.  They swap a t-bond or MBS for a bank reserve.  This doesn’t change the net financial asset position of the private sector.  The bank literally has the same capital position it did before this policy was enacted.  So, the bank can’t create more inside money than it could have before.  And we know that this outside money (reserves) doesn’t flood out into the private sector because it is used ONLY by the interbank system.  Anyone who understood this in 2009 (as many of us did) knew that Murphy was wrong because his understanding of the system was wrong.

So, as we’ve seen time and time again, misunderstanding modern banking and money has resulted in very bad predictions.  Unfortunately, I still don’t see many people agreeing on why Murphy and others were wrong.  That’s not progress.



Got a comment or question about this post? Feel free to use the Ask Cullen section, leave a comment in the forum or send me a message on Twitter.

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services.

More Posts - Website

Follow Me:

  • DanH


  • bart

    Inflation is not equal to any variety CPI.

  • Cullen Roche

    Murphy was wrong…even according to Shadowstats whose inflation measure has a curious correlation to….the CPI.

  • whatisgoingon

    They still cling onto the fact that gold is pricing in monetary (and not price) inflation. The CRB index suggests something entirely different.

  • LVG

    Now if we can just get these people to understand MR.

  • bart

    It’s my belief that Shadowstats SGS-CPI adjustments are too high, but they did hit 10% and by that measure, Murphy was right.

    Your link points to Shadowstats Alternate CPI, not the full SGS-CPI.

  • bart

    And so do many other stats. Using gold as an inflation adjustment across the decades is well beyond foolish.

    But it is agreed that it is money, not only per Central Banks etc. but even by the ISO. Just because you can’t but Wheaties with it at Wally World doesn’t mean it isn’t money, as in a “medium of exchange”.

  • Cullen Roche

    No, the one I posted is the 1990 based – the one John posts on the front of his website. You used 1980. Why not use 1800? Then you could say inflation was a million percent or probably whatever you wanted it to look like. Hell, why not move the goal posts on top of a building, change the type of ball we play with to a cinder block and keep calling it the same sport! :-)

  • scharfy

    gotta know how the plumbing works – good post

  • Pierce Inverarity

    What’s your point?

  • Hans

    Who said that banks reserves are no constraint on borrowing ?

    Could this be the reason for no double digit inflation ?

    Industry Capacity Utilization Rates Chart:

  • bart

    Call it whatever you like, the SGS-CPI was over 10%. It covers the period since the birth of the OER (Owners Equivalent Rent) in 1982.
    Everybody knows of course that the OER is as close as possible to reflecting real house prices, tracks all housing indexes perfectly, and even borders on being almost as good as Mary Poppins… /sarc

    We could also always just use the numbers from the 1930s, double them (or perhaps use a cube root and then add in some Wiemar numbers, but definitely some good single malt and a ceegar [no Monica please]) and then call everything in über deflation if it floats your raft?

    Happy New Year, and party hearty… and we’ll hold each others feet to the fire & brimstone again next year…

  • Cullen Roche

    Ha. Yes. You too Bart. Thanks for the healthy debate this year and best to you and yours in 2013!

  • bart

    What don’t you understand about what I wrote?

  • Sealander

    When the economy is good (i.e. lots of credit-worthy borrowers), banks really are reserve constrained, though, right? In that case, would ballooning bank reserves facilitate the creation of ‘inside’ money and lead to high inflation?

  • Pierce Inverarity

    I understand what you wrote, I don’t understand your point in bringing it up. As Cullen has said, you move the goalposts all the time to fit your agenda, then say that we’re not scoring any points.

  • Cullen Roche

    No, banks are never reserve constrained. The idea that banks are temporarily constrained is Krugman’s liquidity trap. It’s not correct.

    The reality is that there is little demand for loans because of the balance sheet recession.

  • Johnny Evers

    There is clearly something wrong with the current lending mechanism in the economy. Is there some kind of formula that measures the economic growth generated by a loan? For example, a farmer borrows money, clears pasture to run more cattle, hires a laborer, produces more mile, pays the loan back — that would be productive. Borrowing money to buy a new car — maybe not so productive.
    Our government is borrowing money to carry the boomers through retirement, which is defensible as public policy but it won’t grow the economy.
    Meanwhile, that debt will grow — $16 trillion now, on its way to twice GDP within 10 years.
    Clearly we haven’t had inflation, but surely this can’t be a good situation.
    Perhaps the inflation will come when all those financial assets are spent. Or maybe we’ll just muddle along permanantly on the present path, in which the standard of living declines for everybody except those in the financial sector.

  • bart

    The facts are simple about gold being money.
    No goalposts, no moving, just raw facts.
    No points, yet again, for noise without signal.

    Feel free to reject truths and facts as much and as many times as you want to or desire, in order to fit into your carefully constructed bogus castles and ideologies in the sky.

    We continue to be quite amused.

  • bart

    Krugman should be replaced by a button.

    You should be proud of me – whenever I post on his blog (even about simple things like him using public debt instead of total debt figures), I’m 100% on it getting deleted.

  • Sam A

    I don’t get why this concept is so hard for people to grasp. Banks create money by way of making loans, based on the demand for credit. No incremental demand, no new money in circulation. Banks are constrained by their capital ratios, not their reserves. More reserves does not equate to more money in circulation, hence no inflationary threat.

  • Johnny Evers

    You can get people to understand that very easily; however, most people leap to the next step, in which the loan gets paid back, thus removing the money. So in that sense, the loan does not create money, but economic activity generated while the loan is out creates money.

  • Andrew P

    That is what I would bet on. A long term muddle along the present path until something big changes. To get a new growth spurt, we either need a 4th industrial revolution, or the 3rd industrial revolution has to get a big second leg up. The latter could happen if there are major breakthroughs in machine intelligence and robotics. If we muddle along until major resource depletion causes negative growth, well that is more a topic for a doom blog.

  • Andrew P

    Long term deflationary – you are describing Japan to a T.

  • hangemhi

    Using recent info to project debt levels in 10 yrs is wrong. That’s what Clinton did projecting zero debt. As for productive vs non, that is the debate the country should be having.

  • hangemhi

    And if you use a loan to build a house, then pay that loan back when you sell to a home owner who gets an even bigger loan you’ve created wealth that allows for ever bigger loans. And inflation of the asset and your wages leads to bigger loans still and total inside money grows and grows.

  • Alberto

    Inflation or deflation ? Well both of them will kill the stock market as so well explained by Ed Easterling in his books and essays. The FED (and all the others CB) doesn’t want to kill the stock market so we will not see inflation coming from that side. This doesn’t mean inflation will not come, but without a huge fiscal bonus, i.e a real helicopter money drop, we will not see inflation. On the long term, a higher inflation will come from the rising cost of energy and other basic resources, but in the next months we will probably see dropping oil prices and not the contrary. But people who know the physical constraints will just have a huge opportunity to make some easy money. I really hope to see oil at 60$ again.

  • Johnny Evers

    That model is failing in so many way.
    The money the government borrows is not repaid, so there is no multiplier. And while we have asset inflation — benefitting the financial commmunity, primarily, we don’t have wage inflation.
    We’re putting more money in the system and calling it growth.

  • Johnny Evers

    Any pension or defined benefits manager can easily show me the path of government spending the next 20 years.

  • Hans

    Thank you, Sealander and Mr Roche.

  • Jonathan M.F. Catalán

    Who uses gold as a medium of exchange?

  • Sealander

    Ok. But then in what sense is having reserve requirements meaningful? What would happen if they were eliminated? Nothing? I ask sincerely. Thanks for your first response.

  • Cullen Roche

    Pretty much. Many systems like Canada’s don’t even have reserve requirements. A well run banking system can handle a small amount of reserves mainly for interbank settlement processing.

  • lester

    Why do central banks hold it?

  • Dan M.

    Central Banks hold Treasury securities and MBS’s, too. Nobody’s saying it’s not of value… heck, I’m not saying it’s not money. However, barely anyone uses it as a medium of exchange. In fact, there are shops all over dedicated solely to giving people “cash for gold” so they can pay other people with stuff they’ll actually accept.

    I almost think the “moneyness” of assets is less and less important as we could probably figure out ways of using shares of the S&P 500 as money in this electronic age without much hassle.

  • Dan M.


    I still have to see you successfully counter the idea that reserves aren’t at least SOMEWHAT of a counstraint. If reserves aren’t a constraint, why do banks pay us interest? They’d just make loans and pay zero interest on deposits if people so chose to keep their money in that bank.

    If a single bank approaches breaching their reserve requirement, they have to borrow from other banks or the fed, at interest. That is a constraint. The degree to which it IS a constraint depends on the interest rate their being charged vs what they can get from the market, but it’s certainly easier to lend something you have than it is to lend something you have to pay interest to someone else to get first.

    Is it not?

  • Pete

    Hey Jonathan, ever heard of Iran? When the U.S. spread its military complex to its financial transactions, Iran turned to gold for international trade.

  • Cullen Roche

    Well, reserves ARE somewhat of a constraint. After all, they’re assets of the bank. And banks are capital constrained so you can’t just say that banks are capital constrained and then say reserves don’t matter. I say reserves don’t matter in the sense of the traditional money multiplier idea. Banks don’t lend their reserves and their lending is not constrained by their reserve balances. After all, Canada doesn’t even have reserve requirements….

  • Matt

    That’s fine as long as there’s no taxpayer funded bailouts through the treasury or the fed. Do away with federal deposit insurance as well and allow the banks to set their own reserve amount based on the risk they are willing to take. In addition, there would have to be accountability for shareholders and management like there has been in the past.

  • joe

    no, the liquidity trap is the idea that unemployment is so high, the natural rate of interest that clears labor markets is below 0, and the zero lower bound is binding (can’t go below 0).

    this means there is no difference between holding cash and owning a 0% bond. there is no incentive for agents to demand bonds over cash, and there is no incentive for banks to lend. why would you lend someone at a 0% interest rate when the only significant move the price of the asset can take is down?

    this renders manipulation of short-term rates – traditional monetary policy techniques – useless: they can’t go below 0. the Fed can buy all the bonds it wants, but, as long as the unemployment level remains high, short term rates will stay at 0%, and banks won’t lend and investors won’t buy. if the Fed can’t convince investors that inflation will be higher in the future – signaling positive economic growth, falling unemployment – rates will not rise, lending will not pick up and liquidity will just sit in current asset accounts of firms and households, completely unproductive.

    this is exactly what’s happened the last 4 years. despite multiple rounds of large-scale asset purchases by the Fed, unemployment stayed high, inflation stayed low, and rates stayed at 0. liquidity is trapped. it can’t get out into the economy to be used productively.

    oh, and Canada does have reserve requirements. it’s just not called a requirement. it’s called “The Capital Adequacy Guideline”, which means it’s not an enforced law, but it is implied that all banks are expected to comply with the guideline.

  • Tom Brown

    “Capital Adequacy” sounds like a capital requirement (like Basel accords)… not like a reserve requirement. You calculate a “Capital Adequacy Ratio” (CAR) to determine if you’re meeting the capital requirements of the Basel Accords (I, II, and III).

  • joe

    “banks don’t lend their reserves. Banks lend based on their solvency or capital constraint. Reserves are merely an asset of the bank.”

    Banks lend based on their deposits. the reserve requirement is the percentage of deposits the bank must keep at the Fed to cover potential cash withdrawals by depositors. that’s why bank runs are so dangerous. when everyone tries to get their money out at the same time, banks become insolvent.

    the 2008 crisis was particularly interesting because the bank run didn’t happen in the traditional way. it happened in the repo markets, and so investors did not ever witness people lining up down the street to withdraw their money, like you would have seen during the Great Depression or the banking panics of the 19th century. there was never a visible, palpable panic for everyone to see. this is significant when you think about the foreclosure issues that are rampant right now: by not seeing the crisis, the psychology of market participants wasn’t changed dramatically enough to make debt forgiveness more acceptable. see this interview with Gary Gorton

    “When the Fed implements monetary policy like QE they don’t change the capital position of the banks. They swap a t-bond or MBS for a bank reserve. This doesn’t change the net financial asset position of the private sector. The bank literally has the same capital position it did before this policy was enacted.”

    this doesn’t make any sense: what’s the point of monetary policy if the capital position of banks isn’t changed so they can keep lending and credit markets don’t freeze? wasn’t the big issue in the beginning of 2009 recapitalization of the banks to jump start credit markets again?

    “So, the bank can’t create more inside money than it could have before. And we know that this outside money (reserves) doesn’t flood out into the private sector because it is used ONLY by the interbank system.”

    again, this doesn’t make any sense. if rates are 15% when a recession hits, the Fed steps in and buys up short term bonds like crazy, causing the rate to drop, say 10%. so now banks have a boat-load of fresh new cash from the Fed, and they can lend at 5%. why doesn’t this money make it’s way into the economy at this point? why did the Fed buy up everything?

    now replace that with what actually happened in 2008: short term rates went from 4% to 0%. I can understand why 0% loans wouldn’t be made by banks: it’s all downside with little return – the price is most likely only going to fall. in your model, i don’t understand why higher than 0% loans would not be made.

  • joe

    and what do you think a capital requirement is? not the capital required to have on hand in case of withdrawals?

  • joe

    Capital requirement (also known as Regulatory capital or Capital adequacy) is the amount of capital a bank or other financial institution has to hold by its financial regulator. This is in the context of fractional reserve banking and is usually expressed as a capital adequacy ratio of liquid assets that must be held compared to the amount of money that is lent out. These requirements are put into place to ensure that these institutions are not participating or holding investments that increase the risk of default and that they have enough capital to sustain operating losses while still honoring withdrawals.

  • joe

    The reserve requirement is a central bank requirement that stipulates the minimum amount of reserves each bank must hold as a proportion of customer deposits and notes.

  • Cullen Roche

    Banks are in the business of making loans. And they don’t lend at 0%. they lend at a spread over the FFR. As long as there are creditworthy customers entering their doors they’ll lend them money at a reasonable spread above the FFR. The liquidity trap is based on a false understanding of how a bank operates.

    And reserves are an asset for banks. They make up a component of the balance sheet, but don’t represent capital adequacy necessarily.