Money and Inflation, Pt 3 (The Quantity Theory of Money and the Great Inflation)

Here’s my Russ Roberts Podcast.  Patrick sent me my AEI presentation and my Larry Kudlow interview (67 minute mark, 3/23/13).

There are two aspects of fiat money that make the supply and demand for a fiat currency differ from the commodity money model:

1.  The government has almost unlimited control over the stock of currency, and can produce currency at near zero cost.

2.  The demand for money becomes unit elastic, in response to changes in the value of money (1/P).

Here’s how the money supply and demand graph looks for fiat currency:

basedemand

Monetary policy has switched from influencing the demand for the MOA (under the gold standard), to a policy of influencing the supply of the MOA (although money demand is also affected by Fed policies.)  The Fed can shift the supply curve to the left or right via open market operations or discount loans.  Thus the “supply curve” is actually a policy tool, representing the quantity of base money.

In the past I used to teach the quantity theory of money with a “helicopter drop” example, but I now see that won’t work—people get the wrong message.  Traditional OMOs also confuse people.  More specifically:

1.  Some Keynesians believe it matters a lot whether the currency is introduced via a “helicopter drop” or OMOs.  The term ‘helicopter drop’ actually refers to combined monetary/fiscal expansion.  Money dropped out of helicopters is sort of like “welfare” payments, and it’s also an expansion of the money stock.  So it’s like paying for a new entitlement program by printing cash and spending it.  But these Keynesians are wrong.  The fiscal effects are utterly trivial as compared to the monetary effect, at least during normal times.  Increasing the base by 0.2% of GDP during normal times is a big deal.  Increasing debt held by the public by 0.2% hardly matters at all.

2.  Some Austrians worry about “Cantillon effects,” which means they think it’s important to consider who gets the money first.  (Although the term also has other meanings). They assume that that lucky group will boost its spending.  Yet the money is not given away, it’s sold at market prices.  So the person getting the money first is not significantly better off, and hence has little incentive to buy more real goods and services.

Both errors share something in common—the notion that money injections matter because “people with more money will spend more.”  But this subtly confuses wealth and money.  In everyday speech we might say; “a billionaire buys a big yacht, because he has lots of money.”  But we really mean he has lots of wealth.  The billionaire might have very little cash.  So if we are truly going to understand the pure effects of monetary injections (without fiscal or Cantillon effects), we have to consider a form of injection that doesn’t appear to make anyone “better off,” so that people would have no obvious reason to go out and buy more stuff.

I’d like to assume that money is injected according to the following formula:  Suppose there are 100 million Americans that get substantial checks from the government each year (more than $200).  These include tax rebates, veterans benefits, unemployment insurance, government worker salaries, Social Security, etc.  A cross section of America.  The Fed wants to increase the base by $20 billion this year.  Have the Treasury pay each of the 100 million Federal check recipients the first $200 due to them in cash, and the rest by check.  In this case people are not getting any additional money, it’s just that some of it comes in the form of cash rather than the usual checks.  If the Fed had decided not to increase the base, the extra $200 would have been paid by check.  That is the essence of monetary policy, with no distracting bells and whistles.

People don’t want to hold that much extra cash, so they’ll get rid of it.  But how?  Obviously not by burning it.  Now we come to the concept that lies at the heart of money/macro—the fallacy of composition.  Individuals can get rid of the cash they don’t want, but society as a whole cannot, at least not in nominal terms.  How do we reconcile that seeming paradox?

Take an example where the Fed doubles the currency stock, from $200 to $400 per capita (shown in the graph.)  How do we reach a new equilibrium?  In the short run prices are sticky, and short term interest rates might fall.  But over time prices will adjust, and the public will reach a new equilibrium where they are happy holding $400 per capita.  How much do prices have to rise for supply to equal demand at the original interest rate?

If we assume that people care about purchasing power rather than nominal quantities, then prices must double, so that the purchasing power of the stock of cash returns to its original level.  Say that people used to hold enough cash to make one week’s worth of purchases ($200.)  Then prices must rise until one week’s worth of purchases costs $400.  In other words prices rise in proportion to the rise in the currency stock.  And that means the demand curve for the MOA is unit elastic.  In contrast, when silver (or gold) was the MOA, the demand for those assets was not unit elastic.  Currency is special.  Its only value is its purchasing power—it has no industrial uses at all (unlike gold.)

The assumption of a unit elastic demand for currency leads to the Quantity Theory of Money.  If you double the money supply, the value of money will fall in half, and the price level will double.  Of course this assumes the demand for money does not change over time.  But money demand does change.  It would be more accurate to say; “a change in the money supply causes the price level to rise in proportion, compared to where it would be if the money supply had not changed.”  But even that’s not quite right because (expected) changes in the value of money can cause changes in the demand for money.  So all we can really say is:

One time changes in the supply of money cause a proportionate rise in the price level in the long run, as compared to where the price level would have been had the money supply not changed.

That’s because one time changes in the money supply probably don’t shift the real demand for money in the long run.  This is a somewhat weaker version of the QTM, but is the most defensible version.  In my view the QTM is most useful when there are large changes in the supply of money, and/or over the very long run.  Especially when there are large changes in the supply of money, year after year, over a very long period of time. In other words, international data over a long period during the global Great Inflation.  Robert Barro’s macro text (4th ed.) has the perfect data set for thinking about the QTM; 83 countries, over roughly 30 years, when inflation rates were very high and varied dramatically from one country to another.  Here are the top 10 and the bottom 10 on the list:

Country     MB growth    RGDP growth    Inflation   Time period

Brazil              77.4%             5.6%                 77.8%        1963-90

Argentina        72.8%             2.1%                 76.0%        1952-90

Bolivia             49.0%            3.3%                  48.0%        1950-89

Peru                49.7%             3.0%                 47.6%        1960-89

Uruguay          42.4%             1.5%                 43.1%         1960-89

Chile               47.3%             3.1%                 42.2%        1960-90

Yugoslavia       38.7%             8.7% (FWIW)     31.7%         1961-89

Zaire               29.8%             2.4%                  30.0%       1963-86

Israel               31.0%             6.7%                 29.4%        1950-90

Sierra Leone     20.7%            3.1%                  21.5%        1963-88

.  .  .

Canada            8.1%              4.2%                  4.6%         1950-90

Austria             7.1%             3.9%                   4.5%         1950-90

Cyprus            10.5%            5.2%                   4.5%         1960-90

Netherlands      6.4%             3.7%                   4.2%         1950-89

U.S.                 5.7%              3.1%                   4.2%        1950-90

Belgium           4.0%             3.3%                   4.1%         1950-89

Malta               9.6%             6.2%                    3.6%        1960-88

Singapore       10.8%            8.1%                     3.6%        1963-89

Switzerland       4.6%             3.1%                   3.2%        1950-90

W. Germany     7.0%             4.1%                    3.0%        1953-90

Homework for today:

Answer the following 5 questions and you’ll understand the QTM:

1.  Does the “eyeball test” provide more support for the QTM in the low or the high inflation countries?  What does this tell us about its actual applicability to each group?  How does its relative applicability to each group depend on which of the definitions of the QTM (discussed above) is used?

2.  In 71 of the 83 countries the money growth rate exceeds inflation, and in 12 the inflation rate exceeds the money growth rate.  Explain why the ratio is so lopsided.

3.  The gap between money growth rates and inflation exceeds 10% in only one of the 83 countries (Libya–not shown.)  Why does the gap rarely exceed 10%?

4.  Do most of the twelve cases where inflation exceeds money growth occur in low or high inflation countries.  Explain why.

5.  Explain what sort of inflation data would better explain the gap: average inflation rates, the change in the inflation rate, or changes in the expected inflation rate.

I’ll answer tomorrow in the next post.  The commenter with the best set of answers gets a gold star.


Tags:

 
 
 

134 Responses to “Money and Inflation, Pt 3 (The Quantity Theory of Money and the Great Inflation)”

  1. Gravatar of Greg Ransom Greg Ransom
    25. March 2013 at 07:40

    To repeat.

    You understand neither Austrian macroeconomics, nor the Cantillon Effect.

    This does *NOT* address Hayek’s monetary economics and macroeconomics.

    Become competent in the economics *THEN* analysis it and talk about it.

    A professor would never let a freshman do this backwards taking up his class time talking endlessly and critically about things he hasn’t done any work to gain minimal scientific competence to discuss.

    Really. Do the work. The engage the science. Otherwise, spare us.

    Scott writes,

    “Some Austrians worry about “Cantillon effects,” which means they think it’s important to consider who gets the money first.”

  2. Gravatar of Greg Ransom Greg Ransom
    25. March 2013 at 07:46

    Scott, to talk seriously about Hayek & economists or journalists using Hayek’s scientific work, you need to engage the monetary economics of finance, banking and money substitute assets — eg endogenous money and the relation of the Federal Reserve to that.

    And then connect that up to the logic of choice across time involving the valuation of heterogeneous and interconnected production goods of alternative lengths, output and kind.

    *NEVER* seen you do that.

  3. Gravatar of Rob Rawlings Rob Rawlings
    25. March 2013 at 09:01

    Scott,

    I have some question on

    “Have the Treasury pay each of the 100 million Federal check recipients the first $200 due to them in cash, and the rest by check. In this case people are not getting any additional money, it’s just that some of it comes in the form of cash rather than the usual checks:

    – If the government had previously been raising the money for the checks via a tax and they continued with the same level of tax even when the $200 per person is paid in newly minted money – will this still be inflationary? (in this case the new money will in effect be used to reduce the govt deficit or lead to a larger balance being held in a treasure acct somewhere).

    – If the govt didn’t print new money but just reduced the tax by the equivalent amount and continued to send out checks won’t that be just as inflationary as in the money printing case ? (not sure this would be legally permitted – but if they did it anyway and just allowed the acct funding the checks to go into deficit).

  4. Gravatar of phil_20686 phil_20686
    25. March 2013 at 09:36

    (1) So the QTM, as I remember it, takes into account the fact that MB increases alongside production increases are not inflationary, so the correlation should be roughly between MB expansion and (RGDP+inflation). It seems to me that that shows an extremely high correlation for both high and low inflation c countries. Still, the correlation obviously looks better for high inflation countries. If you insist on ignoring RGDP growth as per the definition above then its much better for high inflation countries – obviously.

    (2)If you ignore RGDP growth it causes the correlation to break down at low inflation, as it works only when MB growth>>RGDP growth.

    (3)Because RGDP growth rarely exceeds 10%.

    (4)Low, RGDP growth is much more significant relatively when RGDP roughly equal magnitude to inflation.

    (5) None of the above? I mean you could make an argument that there was a persistent gap between expected inflation and actual expansion of the monetary base, but that such a gap should persist in one direction for fifty odd years strains credibility. I would guess that the next biggest effect was structural: changes in capital flow and associated willingness of foreigners to hold your currency. My intuition is that a lot of Europeans held quite a lot of wealth in dollars and Swiss francs through the 1920-1950 period, and slowly sold those holdings and brought the money back as Europe recovered. US and Switzerland stand out has having more inflation than you would guess, and they are the archetypal haven currencies. Some of the others like Zaire and Sierra Leone experienced massive capital flight due to social unrest.

  5. Gravatar of Max Max
    25. March 2013 at 12:18

    What is the demand for money when seignorage is negative?

  6. Gravatar of Georges Georges
    25. March 2013 at 12:29

    Ssumner, I don’t understand why do you equate the Keynesian reasoning with the Austrian one, a helicopter drop is different from OMO precisely because a helicopter drop is increasing people’s wealth contrary to OMO, so this would usually boost AD. And not because of some fallacy of who receives the money first …

    Now obviously keeping in mind the sumner critique this will only be true if the monetary authority is not sterilising this, which can happen for instance on the ZLB as monetary policy becomes less effective(note the term less and not not effective as it would still be effective but less).

    I guess the big debate is whether the MB or the level of interest rates that determines the price level…

    As for the questions:

    1) on the high inflation sample, mb seems correlated with inflation, while on the low inflation sample, it seems correlated with NGDP growth. Low inflation sample makes sense, for the high inflation case to make sense, it means that velocity has been going up, which is I guess explained by a loss of confidence from the people. So when they see high inflation they expect even higher inflation etc …

    2) If we assume positive RGDP growth, you would expect mb to increase more which explains the finding.

    3)for the gap to be that big, either RGDP growth has to be >10% which is rare, or velocity going down by 10% which is quite difficult to happen.

    4) In high inflation countries, this goes in tandem with question 1, in high inflation countries, expected inflation seems to be quite high which is driving velocity up and inflation up.

    5) I would say the expected inflation.

  7. Gravatar of Don Geddis Don Geddis
    25. March 2013 at 14:35

    Love “Homework for today”! But commenters phil and Georges beat me to it, so I don’t have much to add. FWIW, my answers:
    1. (a) High inflation; (b) there’s some additional factor of a couple percent, which becomes less significant as all the numbers get bigger, but more significant if NGDP is low; (c) for all three of the definitions, high inflation is a better match, because you hope that MB growth = NGDP growth, and that is closer in the high inflation countries. But velocity changes, and we don’t know the path of velocity for each country during the period.
    2. If velocity is stable, MB growth = inflation + RGDP growth, so therefore MB growth > inflation. Velocity can either go up or down or be stable. In order for inflation > MB growth, velocity would have to skyrocket, which is unlikely.
    3. With stable velocity (on average, over the long term), it would require RGDP growth > 10%, which is very rare.
    4. (a) high inflation; (b) with big numbers, velocity only needs to rise a little bit, in order for inflation to exceed MB growth. With low inflation, it would require huge velocity growth to get above MB growth (since you need to make up for the RGDP component of NGDP as well).
    5. Changes in inflation expectations. The missing data is velocity, and we need to at least get some indirect data in order to get a hint on what is happening to velocity.

  8. Gravatar of josh josh
    25. March 2013 at 14:43

    The old-school QTM is kind of a self-fulfilling prophecy, right? The idea being: if you consistently and moderately increase M, then V will remain stable over the long run (because NGDP expectations will remain stable), so NGDP will grow consistently and moderately. So market monetarism, if successful in stabilizing NGDP expectations, would collapse into QTM (?)

  9. Gravatar of ssumner ssumner
    25. March 2013 at 16:33

    Rob, Inflation depends on monetary policy, taxes play almost no role. So the helicopter drop is only a tiny bit more inflationary than an OMO.

    George, If there is no sterilization the fiscal stimulus will increase the price level, but the size of the increase will be utterly trivial compared to monetary stimulus–that was my point.

    Josh, Maybe, but that’s not certain.

  10. Gravatar of Steve Steve
    25. March 2013 at 16:41

    Biotech companies, take note! Being a lame duck enables regrowth of missing spinal tissue:

    Bernanke on QE: “Because stronger growth in each economy confers beneficial spillovers to trading partners, these policies are not ‘beggar-thy-neighbor’ but rather are positive-sum, ‘enrich-thy-neighbor’ actions,” he aid.

    http://www.huffingtonpost.com/2013/03/25/ben-bernanke-fed-world-economy_n_2949959.html

  11. Gravatar of Doug M Doug M
    25. March 2013 at 17:02

    1) Both the high-inflation and the low-inflation countries support QTM
    MV = YP
    d(MV) / MV = dM/M + dV/V
    d(YP) / YP = dY/Y + dP/P
    %cange in M + %change in V = %change in P + % change in Y
    if v is nearly constant.
    Money growth = NGDP growth.

    2) Money growth + velocity change = real growth + inflation
    If velocity change is less than real growth (which should be the case most of the time) then Money growth is greater than inflation.

    3) example is not shown… For the differential between money growth and inflation to exceed 10% either RGDP change or velocity change must be extreme.

    4)This relationship:

    %cange in M + %change in V = %change in P + % change in Y

    Is only linear in for moderate movements in M, V, P and Y.

    When changes become sufficiently large in any one of the variables the affect on the error terms is multiplicative.

    5) What gap?

    Money growth is correlated with changes in prices. (iflation is a change in the price level) Changes in inflation would be a second derivative. So of the three choices given Average inflation rate is the only one that is aplicable. Had you offered the level of expected inflation, I may have to consider that.

  12. Gravatar of Rademaker Rademaker
    25. March 2013 at 17:20

    “But these Keynesians are wrong. The fiscal effects are utterly trivial as compared to the monetary effect, at least during normal times. Increasing the base by 0.2% of GDP during normal times is a big deal. Increasing debt held by the public by 0.2% hardly matters at all.”

    I gather your criticism does not apply to endogenous money theorists, since these argue that the monetary effect takes place only on the initiative of the private banking system, i.e. broad money expands before, not after base money does.

    “2. Some Austrians worry about “Cantillon effects,” which means they think it’s important to consider who gets the money first. (Although the term also has other meanings). They assume that that lucky group will boost its spending. Yet the money is not given away, it’s sold at market prices. So the person getting the money first is not significantly better off, and hence has little incentive to buy more real goods and services.”

    The market price is affected by the expectation that a monetary injection will happen. In as far as the Fed’s actions are predictable (which to a certain extent they always are), these will get front-run by the market. So the Fed typically purchases assets at a premium for being predictable and the successful Fed predictors pocket some if not all of the proceeds of seigniorage.

  13. Gravatar of Steve Steve
    25. March 2013 at 17:23

    1. QTM works better in high inflation countries.
    Stable MB growth is a better target in high inflation countries than low inflation countries.
    It doesn’t matter if you are Keynesian or Austrian.

    2. Money growth exceed inflation in most countries because expected nflation and velocity were both falling.

    3. The power of compounding. 10% for 3 decades is over 1600%. You would need massive changes in velocity.

    4. 5 of the 12 cases of inflation exceeding money growth occur in the top ten inflation cases listed. The hot potato effect is powerful stuff; it really drives up velocity.

    5. The change in inflation rate. High inflation expectations increase velocity, and low inflation expectations lower velocity. The hot potato, again.

    When do I get my gold bar?

  14. Gravatar of Steve Steve
    25. March 2013 at 17:24

    5. I meant change in inflation expectations.

    And dang it, it’s a gold STAR, not a gold BAR! Oh well.

  15. Gravatar of Rademaker Rademaker
    25. March 2013 at 17:40

    “Both errors share something in common””the notion that money injections matter because “people with more money will spend more.” But this subtly confuses wealth and money. In everyday speech we might say; “a billionaire buys a big yacht, because he has lots of money.” But we really mean he has lots of wealth. The billionaire might have very little cash. So if we are truly going to understand the pure effects of monetary injections (without fiscal or Cantillon effects), we have to consider a form of injection that doesn’t appear to make anyone “better off,” so that people would have no obvious reason to go out and buy more stuff.”

    Scott, I understand that under the Fed’s “normal” strategy, it injects solely liquidity and not capital into the market by doing monetary injections. But things are markedly different when highly unconventional circumstances and actions like the current ones are in question.

    When the Fed first buys up government debt and then rips up the paper, you can not argue that the government is no better off than before the Fed’s intervention. It’s debts have been forgiven. That is capital injection, not liquidity injection.

    My suspicion is that the extreme debt growth prior to the 2008 crisis has given rise to a situation where the Fed will end up doing something equivalent to ripping up debt in order to counter debt-deflationary pressure. It has to get rid of debt to counter deflation, so it will have to rip up debt paper or do something equivalent. Most likely it will keep its balance sheet in its expanded state forever. It will roll over debts on the balance sheet perpetually and return any interest on it to the government. This is equivalent to ripping up the paper. It is capital injection, not liquidity injection.

    The “monetary effects” you are hoping for will never be brought about by monetary injections, because they have already occurred in anticipation of the injections, prior to these, not as a result of these.

  16. Gravatar of ssumner ssumner
    25. March 2013 at 17:55

    Steve, It’s actually six of the 12 cases in the top 10, and 8 of the 12 in the top 13 countries.

    Rademaker, I mostly disagree. The government pockets the seignorage, not Fed predictors. And I don’t expect the Fed to be ripping up any paper, except worn out dollar bills.

    Answers tomorrow.

  17. Gravatar of Dan Dan
    25. March 2013 at 18:02

    Hi Scott,

    I listened to the AEI presentation and it seemed you implied that Australia follows NGPD targeting. It was only after listening to Econtalk that I understand you to mean that Australia did well in the Global Financial Crisis because luckily (i.e. China needs coal and iron ore) it had a higher NGDP, rather than because it targeted NGDP. Is this correct?

    I thought you might like Glenn Stevens (RBA Governor) response after being questioned on NGDP targeting:
    “I think I can claim that I did more than most people to build the current framework. I think it works very well and I would be very loath to move off it.”

    As a worldwide authority on NGDP targeting does this seem a reasonable response that will affect the lives of millions of people? :)

    Glenn Stevens later says: “in a case where we have, say, very large terms of trade swings that make nominal GDP rise 10 per cent one year and go down 10 the next year we would have to have a way of figuring out how to handle that if we were targeting nominal GDP”
    I only completed second year economics so I am having trouble understanding fully what he means. Do you think very large terms of trade swings is a good reason to not use a NGDP target?

    Thanks for the blog. I am always learning when I come here.

  18. Gravatar of TallDave TallDave
    25. March 2013 at 18:07

    Thanks Scott, really enjoyed this post, very informative.

  19. Gravatar of Steve Steve
    25. March 2013 at 18:13

    “It’s actually six of the 12 cases in the top 10”

    I count five, but maybe I’m missing something.

    Brazil 77.4% 5.6% 77.8% 1963-90
    Argentina 72.8% 2.1% 76.0% 1952-90
    Uruguay 42.4% 1.5% 43.1% 1960-89
    Zaire 29.8% 2.4% 30.0% 1963-86
    Sierra Leone 20.7% 3.1% 21.5% 1963-88

  20. Gravatar of Rob Rawlings Rob Rawlings
    25. March 2013 at 18:26

    “Rob, Inflation depends on monetary policy, taxes play almost no role. So the helicopter drop is only a tiny bit more inflationary than an OMO.”

    My question relates to why switching a regular payment from check to cash will be inflationary at all , even if the cash is newly printed.

    If base money is defined as bank reserves + paper money then the new cash will increase the money supply , but apart from the small degree to which holding cash rather than having money in the bank increases the propensity to spend then the result of your example will be a decrease in V to match the increase in M , won’t it?

    So unless the govt reduces taxes as a result of this policy or spends the money it saves on something else it will not have a major effect on the price level.

    Am I missing something ?

  21. Gravatar of TallDave TallDave
    25. March 2013 at 18:30

    All right, I’ll take a stab :)

    1) High. The numbers are large and often strongly correlated, almost eerily so.

    2) Productivity gains are deflationary. This ratio may also be affected by population changes over time.

    3) All else being equal, they should approach each other over time, assuming inflation is being calculated correctly and velocity varies around a mean.

    4) I pasted these into Excel, created a column for the difference, and sorted lowest to highest. 5 of the 6 were high inflation.

    5) Expectations uber alles.

    6) Isn’t a little inconsistent of you to support the gold star standard? :)

  22. Gravatar of Fed Up Fed Up
    25. March 2013 at 20:40

    “2. The demand for money becomes unit elastic, in response to changes in the value of money (1/P).”

    I see prices quoted in terms of currency and demand deposits and as long as there is 1 to 1 convertibility, MOA = MOE = currency plus demand deposits.

    Let’s try it this way. $800 billion in currency, $200 billion in central bank reserves, and $6.2 trillion in demand deposits. Next, demand deposits go to $13.2 trillion. The others stay the same. What is the most likely scenario for prices?

    Next, nobody wants currency, so there is $0 in currency, $200 billion in central bank reserves, and $14.0 trillion in demand deposits. I can see very little change in prices even though currency is $0 because MOA = MOE still = $14.0 trillion.

    The point is the M in MV = PY should be MOA = MOE = currency plus demand deposits.

  23. Gravatar of Georges Georges
    25. March 2013 at 23:15

    Rob, I believe this is biggest monetarist point, the hot potato effect. If someone holds cash in his hand, he wants to get rid of it like if you we’re holding a hot potato, so you will want to give it away driving inflation up.

    Scott, this series of posts are very informative, thanks!

  24. Gravatar of Asco Asco
    26. March 2013 at 00:17

    1a. The eyeball test provides more support in the high inflation countries, at least
    if we just look at the monetary base and inflation.

    1b. It works very well in both groups, however it seems more applicable to the high inflation countries – there is no discussion whether the inflation in those countries were caused by high base growth.

    2. Efficiency gains lower costs for a given base, and population growth increases total real money demand. This means that real gdp growth will reduce the price level holding the base constant. For low inflation countries this effect is relatively large, and even in high inflation countries there will be significant real growth over such a long period, making the distribution skewed.

    3.
    Because RGDP growth rarely exceeds 10%, and real money demand is usually pretty stable in the long run. Also, the Taylor approximation for growth rates break down.

    4.
    Inflation exceeds money growth in none of the low inflation countries shown, and in 5 of 10 of the high inflation countries. Because of RGDP growth, money demand has to decrease on top of the increase in money supply for inflation to outpace base growth. In low inflation countries this effect has to be extremely large. For high inflation countries this can happen because:

    – The RGDP effect is relatively small

    – Money demand becomes unstable at high inflation rates. The opportunity cost of holding money (the nominal interest rate) increases 1 to 1 with expected inflation (or NGDP, but the distinction is not important for high inflation countries). In an inflationary environment, expected inflation will be very high as well, and very unpredictable.

    Money demand decreases because of the high opportunity cost of holding money, and great uncertainty about future inflation will reduce money demand if people are risk averse.

    5)
    Aren’t these already average inflation rates?

    Increases in the expected inflation rate will reduce money demand immediately and raise prices gradually, but the effect on money demand will die out if it’s a one time increase.
    For high inflation countries, there may not be much trust that it will be a one time increase though.

    If this is just about compounding then, eh, I don’t know.

  25. Gravatar of Bill Woolsey Bill Woolsey
    26. March 2013 at 03:15

    Scott:

    Assume a balanced budget. Government collect taxes in the form of checkable deposits. The government then spends those funds by writing checks.

    Now, the government makes some of its payments with newly-printed currency.

    It is still collecting the same taxes, so the government ends up holding a balance in its checkable deposit. This is the tax revenue it didn’t spend. That is, the spending that was instead funded by currency creation.

    This represents lending by the government to the banks. The creation of currency, even though there was no budget deficit at all, involves lending by the goverment to the banks.

    If the government is running a budget deficit and selling debt to fund some government spending (the normal situation these days,) then when the government makes expenditures with newly created currency to fund some of the expenditures, the natural assumtion is the government sells less debt–borrows less.

  26. Gravatar of Justin Irving Justin Irving
    26. March 2013 at 05:15

    Scott, you should team up with one of these free online universities (like Coursera)and do some monetary theory lectures. You know, in your free time.

  27. Gravatar of dtoh dtoh
    26. March 2013 at 05:40

    Scott,
    I think you have the same problem here that you do in explaining NGDP growth through money issuance. What is the transmission mechanism?

    This is actually a hot topic in Japan at the moment among both politicians and business people.

  28. Gravatar of ssumner ssumner
    26. March 2013 at 05:58

    Dan, Yes, if there were very large terms of trade swings you might want to deviate from strict NGDP targeting. But I do attribute Australia’s success mostly to keeping NGDP growing along a fairly stable trend line, not the commodity boom.

    Steve, You are right! Barro’s book has a typo.

    Rob, Why would your personal demand to hold cash rise just because the Fed increased the supply of cash? If not, then prices must rise until the real quantity of cash hasn’t changed.

    Fed up, I prefer the base, as it is the aggregate actually controlled by the Fed. Changes in the demand for demand deposits matter only to the extent that they impact the base, or whatever the medium of account is.

    BTW, what are Cyriot DDs worth? The MOA must have a price that doesn’t change.

    Thanks Georges.

    Bill, That sounds right.

    Justin, That’s what I’m working on.

    dtoh, The transmission mechanism is clearly the hot potato effect, as we are looking at the long run, and real asset prices are not affected in the long run. The short run transmission mechanism may differ, as you claim.

  29. Gravatar of Rob Rawlings Rob Rawlings
    26. March 2013 at 06:12

    “Rob, Why would your personal demand to hold cash rise just because the Fed increased the supply of cash? If not, then prices must rise until the real quantity of cash hasn’t changed.”

    I think I’m trying (apparently not very well) to say something similar to Bill.

    The point is that the recipients of the cash are no better off than before as a result of the changed composition of their income so will not increase their spending much. If they don’t want to increase their cash holdings they will just deposit the cash at the bank and the situation will be the same as if they just got a check.

    Increased spending will rather be driven by the banks lending out some of the reserves they now have (if the govt simply leaves it in an account at the bank) or by the people who would otherwise be lending money to the govt using that money in other ways (if the govt deficit gets reduced)

  30. Gravatar of Rob Rawlings Rob Rawlings
    26. March 2013 at 06:33

    Or by the people who now pay less tax if the govt reduces tax as a result of the new money creation.

  31. Gravatar of Mike Sax Mike Sax
    26. March 2013 at 08:29

    Scott I agree these series of posts are highly informative. However, I second Rob Rawlings question.

    I must be missing what he’s missing.

  32. Gravatar of Peter N Peter N
    26. March 2013 at 09:46

    I agree. If the government sends me $200 in cash or a $200 checkand then collects $200 in taxes, the kind of money I briefly had makes no difference. The bank adjusts it’s cash on hand preferences accordingly and it’s exactly as it was before.

    If OTOH the Fed buys a bond it is exchanging a more liquid asset for a less liquid one, and there’s no reflux path unless the Fed decides to start selling.

    Now if the Fed prints $200 and puts it in the envelope with my $200 check, that’s different. After I pay my $200 in taxes, that $200 is still somewhere in the private sector providing liquidity.

    Of course, the Fed could then sell a $200 bond and sterilize the process, but why should they, given they gave the $200 to me in the first place.

    That’s a helicopter drop, and it adds the most monetary services, since I didn’t have to give up some instrument with an inferior but still valuable liquidity premium to get it. And, of course, there’s a wealth effect and a debt paydown (paying off excess debt has a higher utility than acquiring additional wealth, since prevention of disutility = utility) effect in addition. We can argue about how significant they are.

  33. Gravatar of Geoff Geoff
    26. March 2013 at 12:23

    Dr. Sumner:

    “Some Austrians worry about “Cantillon effects,” which means they think it’s important to consider who gets the money first. (Although the term also has other meanings). They assume that that lucky group will boost its spending. Yet the money is not given away, it’s sold at market prices. So the person getting the money first is not significantly better off, and hence has little incentive to buy more real goods and services.”

    Sorry, but this is incorrect. The money “sold” is not in fact sold at “market” prices. If it were sold at market prices, then OMOs would have no additional affect on spending or prices in the economy, that an absence of said OMOs would have generated. A clear absurdity.

    The error is conflating prices that prevail with a monopoly in money and inflation, and prices that prevail without a monopoly in money and inflation. The latter generates different prices and price levels than the former.

    The confusion over whether money is “given away” as opposed to “sold” misses the mark entirely. That is not the right question to ask. The right question to ask is whether prices with inflation from central banks is different from prices without inflation from central banks. The answer to this should be trivial.

    Now, as regarding the Cantillon Effect, since we know that inflation from central banks affect spending and prices, the next question to ask is whether or not inflation’s affect on prices are homogeneous or non-homogeneous. The first question’s answer can help here. If we know that inflation from central banks do generate spending and prices different from what would otherwise prevail without inflation or central banks, and we also take into account the fact that primary dealer banks and institutions are the SOLE go-between, the connection, between central banks and the rest of the economy, then if we know that central banks affect spending and prices throughout the economy in ways that are different than what would prevail without inflation from central banks, then it MUST be the case, it HAS to be the case, it is LOGICALLY NECESSARY, that inflation from central banks absolutely without a doubt must generate a different trajectory of spending and prices at the initial inflation injection points, than otherwise would have been the case had the central banks not inflated at all, or inflated at different points in the economy.

    In other words, if inflation from central banks does not generate a different trajectory of spending and price at the initial injection points that what would otherwise prevail without inflation, then it is logically necessary that those initial injection points (primary dealers, etc) must not generate a different affect on the spending or selling prices of those they trade with in the secondary markets, and by extension, those secondary markets then cannot bring about a different trajectory of spending and prices of those they trade with in the tertiary markets, and so on down the line, throughout the entire economy. Since we know that is nonsense, since we know that the Fed can bring about different spending and price trajectories throughout the economy, and the central only deals with the primary dealers, then it is necessary that the spending and prices at the initial injection points must be different from what otherwise would have prevailed without inflation.

    The primary dealers do NOT buy new dollars at “market” spending values or prices. They buy them at different spending values and prices, so that their trade partners’ spending and price setting is affected, which is necessary for the central bank to affect spending and prices throughout the economy.

    In basic logical form:

    If C is only dealing with P, and we know that C is bringing about different spending and prices in M, then it must be the case that P is changing spending and prices in M, courtesy of C, and that C must be changing spending and prices at P.

    There is no denying this without denying that C affects M. If you deny C affects P, then you are making it impossible for P to affect M, and thus you are making it impossible for C to affect M, an absurdity.

    —————

    If economic logic doesn’t convince you, then just ask why are the primary dealers interested in being primary dealers? Why are so many banks and institutions desiring to be primary dealers? Why aren’t the existing primary dealers doing whatever they can to get off the primary dealer list? Clearly by virtue of the actions of the individuals in the primary dealer institutions, there is a significant advantage, a gain, to be made by being a primary dealer, or else the individuals involved would not bother.

    I recall you once saying that receiving the new money first actually HARMS the initial receiver’s interests. The fact that your reasoning is taking you to a fallacious conclusion, SHOULD be an incentive, an encouragement, for you to seriously rethink your position.

  34. Gravatar of Georges Georges
    26. March 2013 at 13:05

    Geoff:

    “Sorry, but this is incorrect. The money “sold” is not in fact sold at “market” prices. If it were sold at market prices, then OMOs would have no additional affect on spending or prices in the economy, that an absence of said OMOs would have generated. A clear absurdity.

    Hein????
    So the primary dealers are getting a discount??? Really??
    And then your second statement OMOs have no impact on spending or prices, hein???

    The only argument where this would be true is if people had no value whatsoever for money (a cashless society), but even proponents of such theories (such as Michael woodford) acknowledge that this is not the realistic case. Let’s say we are away from the ZLB how do you think the fed is setting the interest rate? A hint using OMO, so clearly OMOs have an impact on prices even if they are done at market prices!! You can argue whatever you want at the ZLB about the efficacy of OMOs but they are clearly done at market price still!

  35. Gravatar of Peter N Peter N
    26. March 2013 at 13:32

    “so the primary dealers are getting a discount??? Really??”

    Yes, in effect, they are. They profit, among other ways, from information asymmetry. All bond dealers do, but primary broker dealers have an edge.

    http://dealbreaker.com/2013/01/federal-prosecutors-dont-appreciate-former-jefferies-traders-vivid-imagination/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+dealbreaker+%28Dealbreaker%29

  36. Gravatar of Geoff Geoff
    26. March 2013 at 13:48

    Georges

    “Hein???? So the primary dealers are getting a discount??? Really?? And then your second statement OMOs have no impact on spending or prices, hein???”

    No. I didn’t say OMOs have no impact on spending or prices. Dr. Sumner is making that claim.

    “The only argument where this would be true is if people had no value whatsoever for money (a cashless society), but even proponents of such theories (such as Michael woodford) acknowledge that this is not the realistic case. Let’s say we are away from the ZLB how do you think the fed is setting the interest rate? A hint using OMO, so clearly OMOs have an impact on prices even if they are done at market prices!! You can argue whatever you want at the ZLB about the efficacy of OMOs but they are clearly done at market price still!””

    To the extent this passage assumes I believe OMOs have no affect on spending or prices (specifically the spending and prices concerning that which the primary dealers buy), it is not related to what I said. I am trying to see how it would relate even if it doesn’t make that assumption, but I am at a loss…

  37. Gravatar of Georges Georges
    26. March 2013 at 14:32

    Geoff, yes this passage assumes that indeed. If I read back your message you are saying that if OMOs were done not at market price then they would have no impact on prices and spending, right?
    It’s this assumption that I’m contesting.

    Peter N, what are you trying to say with this article? That there are bond traders committing frauds? You bet there are! But I don’t see how can this be related to how monetary policy is done, and I would hardly be convinced that OMOs are only working if there is a fraud involved!

    Primary dealers are getting commissions from those transactions no doubt, but this is tiny compared to the size of those operations and is certainly not the reason why OMOs impact prices.

  38. Gravatar of Georges Georges
    26. March 2013 at 14:42

    Geoff, what prof sumner is saying is that OMOs are not increasing the wealth of the people (which is certainly true at least on aggregate) and so is not changing the propensity to spend of the people, but it’s increasing the monetary base which would then have inflationary impact which would make people spend more.

  39. Gravatar of Georges Georges
    26. March 2013 at 14:42

    Geoff, what prof sumner is saying is that OMOs are not increasing the wealth of the people (which is certainly true at least on aggregate) and so is not changing the propensity to spend of the people, but it’s increasing the monetary base which would then have inflationary impact which would make people spend more.

  40. Gravatar of Georges Georges
    26. March 2013 at 14:43

    Geoff, what prof sumner is saying is that OMOs are not increasing the wealth of the people (which is certainly true at least on aggregate) and so is not changing the propensity to spend of the people, but it’s increasing the monetary base which would then have inflationary impact which would make people spend more.

  41. Gravatar of Georges Georges
    26. March 2013 at 14:43

    Geoff, what prof sumner is saying is that OMOs are not increasing the wealth of the people (which is certainly true at least on aggregate) and so is not changing the propensity to spend of the people, but it’s increasing the monetary base which would then have inflationary impact which would make people spend more.

  42. Gravatar of Georges Georges
    26. March 2013 at 14:43

    Geoff, what prof sumner is saying is that OMOs are not increasing the wealth of the people (which is certainly true at least on aggregate) and so is not changing the propensity to spend of the people, but it’s increasing the monetary base which would then have inflationary impact which would make people spend more.

  43. Gravatar of Georges Georges
    26. March 2013 at 14:46

    Sorry for the spam, not sure what happened here it posted 5 times …

  44. Gravatar of Geoff Geoff
    26. March 2013 at 14:59

    Georges:

    “Geoff, yes this passage assumes that indeed. If I read back your message you are saying that if OMOs were done not at market price then they would have no impact on prices and spending, right?”

    No, you still don’t have it. I am saying that if OMOs were done not at market prices and spending (i.e., meaning at non-market market prices and spending), then they would have an impact on what would otherwise be market prices and spending.

    In other words, if OMOs are done at market prices and spending, then they would not have any affect on spending and prices over time throughout the economy. The fact that OMOs do affect prices and spending throughout the economy, means that it is necessary that for the only institutions the central bank does deal with, the primary dealers, the prices and spending at this initial stage MUST be affected. If it isn’t affected, the primary dealers could not affect the spending and prices regarding the exchanges they make with others not the central bank.

    “Geoff, what prof sumner is saying is that OMOs are not increasing the wealth of the people (which is certainly true at least on aggregate) and so is not changing the propensity to spend of the people, but it’s increasing the monetary base which would then have inflationary impact which would make people spend more.”

    I am not saying inflation increases real wealth in the aggregate either. I am saying it increases relative spending and prices for the primary dealers, which is necessary in order for the primary dealers to then increase the spending and prices throughout the rest of the economy. This is because the spending and prices in the rest of the economy, in the aggregate, must be affected by the primary dealers if we are going to argue that central banks affect prices and spending throughout the economy. The central bank only deals with the primary dealers, so if the rest of the economy is affected, the primary dealers must be affected, and by affected I mean different from what would otherwise have prevailed without the inflation from the central bank.

    It is contradictory to hold both of these positions at the same time:

    1. Central bank inflation affects prices and spending throughout the economy.
    2. Spending and prices at the primary dealer stage are not affected by central bank inflation.

    PS You posted 5 times because you clicked submit 5 times because you thought it didn’t submit properly, due to temporary delays, glitches, etc, in the server.

  45. Gravatar of Georges Georges
    26. March 2013 at 15:29

    Geoff,

    Yes I think I got your point and I don’t agree with it. What you are saying is that the fed is buying the bond from the primary dealer with the primary dealer benefiting from this transaction. This will change the market prices of bonds to reflect this transaction (I find that plausible but this would benefit all bond holders in the economy and not just the primary dealers).

    What you are saying is that inflation will necessarily go from one person to the next and cannot kind of propagate to the whole population. I’m saying it can.

    First of all the biggest beneficiaries of this are bond holders and not primary dealers (they are not necessarily bond holders), second OMOs could have impact on people’s expectation (indeed I argue that in the ZLB it’s almost the only impact it has). Third it could have an impact on the interest rates lowering them down so everyone will be impacted by this.

    If OMOs are very efficient then bond prices will instantly drop and indeed bond holders will gain from this but not necessarily primary dealers, it’s not who receive the money that benefits but it’s the holders of the asset that is gaining in value who is gaining. And you also have the expectation effect which would have a more homogeneous impact.

  46. Gravatar of Geoff Geoff
    26. March 2013 at 15:53

    Georges:

    “Yes I think I got your point and I don’t agree with it.”

    You keep disagreeing with contradictory, mutually exclusive “points” that you thought I was making.

    I think you’re just disagreeing for the sake of disagreeing. Why else would you disagree with two exhaustive claims, one after another, despite the fact that you are logically obligated to agree with at least one of them?

    “What you are saying is that the fed is buying the bond from the primary dealer with the primary dealer benefiting from this transaction.”

    If the primary dealer did not benefit from the transaction, then the primary dealer would not engage in the transaction. The fact that they do engage in the transaction suggests they are indeed benefiting.

    Moreover, you’re somewhat missing the mark here. Put aside “who benefits” for a second and ask yourself if it is possible for the central bank to affect spending and prices throughout the economy while not affecting the spending and prices of the only institutions it deals with, the primary dealers.

    “This will change the market prices of bonds to reflect this transaction (I find that plausible but this would benefit all bond holders in the economy and not just the primary dealers).”

    It is not actually relevant to this issue whether or not the security types the central bank buys goes up or down as a class, which would seem to suggest that every holder of said security type gains or loses together.

    For one thing, not everyone can be a speculating bond seller. There must be buyers, that is, there must be those who do not own those bonds but stand to buy those bonds at higher or lower prices. It is a mistake to believe that because prices of bonds the central bank does not specifically buy change due to OMOs, that everyone can in principle take advantage of the gains if they exist, or avoid the losses if they exist.

    If everyone bought bonds hoping to sell them at, say, higher prices, then the prices would collapse to zero, since the demand for bonds would be zero as everyone is intending to sell.

    But this is besides the main issue here. The issue is whether or not inflation affects the spending and prices at the primary dealer stage, compared to what it otherwise would have been without the inflation.

    Dr. Sumner is arguing that the primary dealers are paying “market prices” for money and bonds. My argument is counter-point to that position. My argument is that it is wrong. I explained why I think it is wrong, and so far you have not addressed this, but rather detoured to other, tangential points that don’t directly address the issue.

    “What you are saying is that inflation will necessarily go from one person to the next and cannot kind of propagate to the whole population. I’m saying it can.”

    Well then you’re wrong about that too. If the central bank doesn’t send everyone checks, then inflation necessarily is a “propagating” phenomena.

    “First of all the biggest beneficiaries of this are bond holders and not primary dealers (they are not necessarily bond holders)”

    I am not making a case as to who are the “biggest” beneficiaries. I am solely addressing whether or not spending and prices are affected at the primary dealer stage, due to OMOs.

    “second OMOs could have impact on people’s expectation (indeed I argue that in the ZLB it’s almost the only impact it has).”

    Everyone cannot have the same expectations, and everyone cannot act on the same expectations.

    “Third it could have an impact on the interest rates lowering them down so everyone will be impacted by this.”

    If rates decline, then, ceteris paribus, there must be an abstaining from spending on things other than bonds, to make available the funds with which to buy bonds at the higher prices.

    If this happens, then those who sell bonds are benefited at the expense of those who buy bonds at the higher prices.

    “If OMOs are very efficient then bond prices will instantly drop and indeed bond holders will gain from this but not necessarily primary dealers, it’s not who receive the money that benefits but it’s the holders of the asset that is gaining in value who is gaining.”

    Nothing in human life is instantaneous.

    “And you also have the expectation effect which would have a more homogeneous impact.”

    Not everyone has the same expectations, nor can they have the same expectations for exchanges to take place. In order for exchanges to take place, there has to be differences in opinions on value.

  47. Gravatar of Peter N Peter N
    26. March 2013 at 16:11

    Primary broker dealers have more price information available to them, so they can earn superior spreads. With no posted prices, the market isn’t efficient. The money involved isn’t trivial, because the volume is huge. Whether this information asymmetric rent has any macro effect, I couldn’t say.

    However I think the banking sector as a whole has an extremely bad effect. Their share of the economy has reached a point not seen since 1932.

  48. Gravatar of Georges Georges
    26. March 2013 at 16:20

    Geoff, let me start with a few questions this way we can understand each other:

    1) OMOs cause inflation?
    2) if answer to 1 was yes, is it inflation caused from the primary dealers being wealthier and therefore spilling over?

    My answers are 1) yes 2) no
    The dealers are getting a commission on their sell and will benefit from the general NGDP growth but this is it, there is no domino effect.

    As for the interest rate story, this is not a zero sum game, if ir goes down, loans are cheaper and people will therefore spend and invest more boosting AD, it’s not a pure substitution effect.

    Rates are at 10%, you will not get a loan and buy this new car you had in mind, now rates are at 0.5% you will take a loan and buy it, I don’t see how can this crowd something else out? You can argue that if you take a loan someone must be borrowing you, yes the government/central bank would be.

    Most people by expecting the fed will keep the rates down for 10Y will take out more loans which will increase AD and generate higher NGDP, you don’t actually even need to do anything, we have seen the market reacting to just the fed announcing some program, this is forward guidance.

    You can see how the stock market reacts almost instantaneously to the Fred’s announcements. You don’t even need OMOs for that matter (although OMOs sure help a lot with the convincing bit and with flooding the market with new loans)

  49. Gravatar of Geoff Geoff
    26. March 2013 at 18:45

    Georges:

    If you accept that the central bank’s inflation results in a change to aggregate spending and prices, and if you accept that the central bank makes exchanges with only the primary dealers, then you are logically obligated to hold that the central bank affects the spending and prices of primary dealers, for if it didn’t, then it could not affect aggregate spending and prices.

    You cannot hold that the central bank causes aggregate inflation, without also holding that the central bank affects the prices of exchanges with the primary dealers. If you deny the latter, then you must deny the former.

    As regards to your opinions on interest rates, they have no bearing on the issue.

    The error you are making is that you are completely evading the necessary middle step between central bank activity, and the subsequent affect on aggregate spending and prices. There is indeed a “spill-over” effect. To deny this is to display a misunderstanding of the nature of inflation.

  50. Gravatar of ssumner ssumner
    26. March 2013 at 19:28

    Rob, You said;

    “If they don’t want to increase their cash holdings they will just deposit the cash at the bank and the situation will be the same as if they just got a check.”

    Sorry, that won’t work either, as the bank won’t want to hold cash that earns no interest.

    Mike and Peter, See my answer to Rob. You don’t want to think about what makes people want to spend more, but rather the fact that people want to get rid of money they don’t want, but they cannot do so (collectively.) Inflation is not fundamentally about prices rising, that’s a side effect. It’s about cash losing value because the Fed has printed more cash than people want to hold.

    Geoff, You are wrong, I suggest taking a look at my earlier posts on the Cantillon effect.

    Georges, Don’t worry, it was worth repeating 5 times. :)

  51. Gravatar of Geoff Geoff
    26. March 2013 at 19:49

    Dr. Sumner:

    “Geoff, You are wrong, I suggest taking a look at my earlier posts on the Cantillon effect.”

    No, you are wrong, and I showed why you are wrong in those earlier posts.

    “Georges, Don’t worry, it was worth repeating 5 times.”

    Georges’ post did not address the issue raised.

  52. Gravatar of Peter N Peter N
    26. March 2013 at 21:13

    “Sorry, that won’t work either, as the bank won’t want to hold cash that earns no interest.”

    Right, so the bank will send its excess cash to the Federal Reserve where it will be credited to its reserve account. Problem solved, the cash has left the system. It’s back where it started, at the Fed.

    That’s what we don’t get. Why doesn’t the bank just send the excess cash back to the Fed in exchange for reserves?

  53. Gravatar of dtoh dtoh
    26. March 2013 at 22:12

    Scott,
    I hate to keep on harping on this, but to achieve a more general understanding and acceptance of the theory, you need to delve more into the transmission mechanism. If you look at the objections raised by politicians, economists, and journalists it all relates back to a failure to understand the transmission mechanism.

    Just to give you an example, please explain how OMP in your model does not result in all inflation and no real growth, or alternatively explain why OMP doesn’t result in no inflation and all real growth, or explain why at the ZLB OMP doesn’t just result in a decrease in V.

  54. Gravatar of Max Max
    26. March 2013 at 22:41

    The concept of a unique quantity of money for a given price level only applies if seignorage (on *both* reserves and currency) is positive. This hasn’t been the case since 2008.

    Note that the Fed could make seignorage on reserves positive by lowering interest on reserves, but then seignorage on currency would turn negative.

  55. Gravatar of Georges Georges
    26. March 2013 at 23:09

    Geoff, my point on interest rate was a response to your post on the rates declining bit.

    I don’t see how it’s contradictory to say that inflation is happening in the economy not through a spillover effect, I gave you one mechanism where this could happen (expectations).

    Another mechanism, bonds are trading at 100, fed announces I want to buy bonds at 110, instantaneously all bonds and close substitutes to bonds will drop in price, this is a generalised inflation caused by not a spillover effect but a mere declaration (Bond traders will only earn commission on this and that’s it which is just a few bps).

    I guess let’s agree to disagree, or as prof sumner has suggested check the cantillon effect post, but please explain one fact: how is the stock market reacting instantaneously to the Fed’s announcements (and I’m saying announcements not even actions) if inflation has to propagate through spillover effect?

  56. Gravatar of Geoff Geoff
    27. March 2013 at 03:57

    Georges:

    “Geoff, my point on interest rate was a response to your post on the rates declining bit.”

    I didn’t say anything about rates declining over time. My position on interest rates is that there are at least two nominal forces acting on interest rates, one that raises rates, and one that decreases rates. The nominal force that raises rates is indirectly through price inflation and the resulting inflation premiums. The nominal force that decreases rates is directly through increasing the supply of credit and the resulting “liquidity effect” negative premium.

    Which one of these two forces dominates the other is an empirical question, but both are always present. The nominal force that decreases interest rates, even if price inflation is putting an increasing pressure on interest rates such that it dominates, is still making interest rates lower than they otherwise would have been had the inflation entered the economy directly in the purchase of final goods.

    This is the counter-factual that relates to the spending and prices that are different with inflation than without inflation, if the initial inflation is exchanged for bonds that we traditionally model by including only one nominal force, namely the inflation premium one, that leads to such statements as “Inflation results in higher interest rates, what are you talking about?”

    Your response is also not addressing my actual point about whether the spending and market prices are different from what they otherwise would have been had the inflation not taken place.

    “I don’t see how it’s contradictory to say that inflation is happening in the economy not through a spillover effect, I gave you one mechanism where this could happen (expectations).”

    That’s unfortunate. To restate it, it’s contradictory because the primary dealers are the sole conduit by which the central bank affects aggregate spending and prices. If the spending and prices at the primary dealer level are no different with the OMOs, then it follows that aggregate spending and prices in the economy will be no different.

    Expectations cannot raise aggregate spending and prices year after year after year. Expectations without an actual corresponding “spill-over” from the inflation, would result in aggregate spending and prices to top out. This follows from the QTM, which holds that the amount of spending that exists in the economy is ultimately a function of how much money actually exists in the economy.

    “Another mechanism, bonds are trading at 100, fed announces I want to buy bonds at 110, instantaneously all bonds and close substitutes to bonds will drop in price, this is a generalised inflation caused by not a spillover effect but a mere declaration (Bond traders will only earn commission on this and that’s it which is just a few bps).”

    This is not a general increase in prices that is a necessary component to the argument being considered. What you are describing here is a rise in spending on bonds which comes at the expense of a decreased spending for other goods/securities. If there is no inflation of the money supply, then spending more on X means there is less to spend on everything else. General prices and general spending do not rise in this way.

    Only if the central bank actually increases the supply of money, can any increased spending on bonds NOT be accompanied by a decreased spending on everything else, such that general spending and general prices rise over time.

    You are placing far too much emphasis on expectations. This is problematic, on multiple levels. One, it is leading you to suppose that price adjustments are instantaneous, which is impossible, two, it is leading you to believe that something only inflation of the money supply can do, can be done by mere words, and three, it is leading you to ignore the real world “spill-over” effect that no amount of expectations and preparation can avoid.

    “I guess let’s agree to disagree, or as prof sumner has suggested check the cantillon effect post”

    I’ve shown how Dr. Sumner is wrong in that post. Why are you blindly repeating his recommendation, when you clearly don’t have all the facts to make such a statement?

    “but please explain one fact: how is the stock market reacting instantaneously to the Fed’s announcements (and I’m saying announcements not even actions) if inflation has to propagate through spillover effect?”

    The stock market is not reacting instantaneously to Fed announcements. Your assumption is wrong.

    Not only does it react with a time lag, but the full effects of an inflation are not immediately priced into stocks. The rise in prices of stocks we see today are the result of past inflation as well as recent inflation, as more and more people whose incomes have finally risen during the “spill over effect” transition, are able to put forth a higher nominal demand for stocks and increase stock prices further. Prices are a function of supply and demand.

    Inflation of the money supply does not all instantly translate into nominal demand for stocks. Stock prices rise as the inflation spreads throughout the economy, raising the nominal incomes of individuals (in a process better described as) “one by one”, and as those individuals save and invest that new money, then stock prices will go up further (if they went up initially).

    There is no way to arbitrage this by you paying higher stock prices initially, because you can’t scientifically predict just how much of the new money from subsequently higher incomes will indeed go to stock price demand.

    MM theory is fundamentally flawed in its denial of “long and variable lags.”

  57. Gravatar of Geoff Geoff
    27. March 2013 at 03:59

    Also, these positions cannot co-exist:

    1. Prices instantly adjust to changes in monetary inflation.
    2. Unemployment is generated by changes in monetary inflation.

  58. Gravatar of Rob Rawlings Rob Rawlings
    27. March 2013 at 05:47

    “Sorry, that won’t work either, as the bank won’t want to hold cash that earns no interest.’

    So I think you are saying: I take the cash I don’t want to hold to the bank and they use it as reserves to lend out and this will increase spending and the price level

    But I am still missing it:

    before:
    I get a check for $1000 and I spend it all by writing checks

    after:
    I get a check for $800 and $200 in newly printed bills. I spend it all by writing checks and spending cash or I take the cash to the bank and write checks just like before.

    I don’t see any real difference between before and after as far as either I or the bank is concerned. I still think the beneficiaries of this new money are the people who are paying me my $1000 as $200 now gets newly created and they can spend $200 on something else.

  59. Gravatar of ssumner ssumner
    27. March 2013 at 06:32

    Peter, You said;

    “Right, so the bank will send its excess cash to the Federal Reserve where it will be credited to its reserve account. Problem solved, the cash has left the system. It’s back where it started, at the Fed.”

    That won’t work either, as the banks don’t want to hold non-interest-bearing reserves at the Fed, when they can hold interest bearing T-bills.

    dtoh, I’ll do that in later posts.

    Max, How can seignorage on currency be negative?

    Rob, You are right that your behavior is initially no different–that was my point about the non-importance of Cantillon affects. You are not richer, and hence don’t buy more goods (initially). But total NGDP begins to rise due to the hot potato effect. Your behavior changes in one respect–you try to get rid of excess cash balances. As an individual you can do so, but collectively society cannot do so. Think in terms of the fallacy of composition, that’s the concept that underlies all of monetary economics. Everything else is just a footnote.

    You are thinking in Keynesian terms, and you CANNOT understand monetary economics while thinking in Keynesian terms. It’s not about spending on real goods and services, it’s about the supply and demand for the monetary base. When the supply goes up, the base loses value, for the same reason that more apples being harvested make apples lose value. Othe things cost more in apple terms, but not because the “aggregate demand for real goods in apple terms” went up. It was because apples were worth less.

  60. Gravatar of Rob Rawlings Rob Rawlings
    27. March 2013 at 07:15

    Scott,

    Surely whenever new money is created someone has to get it and spend it for it to have any effect?

    In the case of OMOs the new money is spent by the fed on assets that it hold hold to implement monetary policy so the effects are more-or-less neutral in its distributional effects.

    In your example the new money is spent by the govt agency that issue the checks (not by the recipients). If they don’t increase spending as a result of the new money but just hold bigger balances then I still contend that the monetary base will increase, but velocity will fall (as a result of the agencies larger balances) and the price level won’t change.

  61. Gravatar of Rob Rawlings Rob Rawlings
    27. March 2013 at 07:17

    By “If they don’t increase spending” I mean the agency that issues the checks that is the beneficiary of the new money.

  62. Gravatar of Peter N Peter N
    27. March 2013 at 08:06

    “That won’t work either, as the banks don’t want to hold non-interest-bearing reserves at the Fed, when they can hold interest bearing T-bills.”

    The two forms of base money are interchangeable. It doesn’t matter which form the government sends me.

    I think your hypothetical intent was:

    Suppose instead of sending ma check for $1000 from the treasury, the check is for $800, and the Fed supplies the last $200 with money from it’s own balance sheet, not the treasury’s account.

    Now you’ve increased the monetary base by $200. However, it doesn’t matter what form of money is sent, if it comes from the treasury’s account, because forms of base money are fungible, and banks can invest excess reserves in t-bills it they want.

    Interest on reserves gives banks an alternative to T-bills. Otherwise there would be a shortage of T-bills, and money market funds would have a problem staying solvent.

    I’m using this reference:

    http://synthenomics.blogspot.com/2012/08/interest-on-excess-reserves-illustrated.html

    which itself draws on Dave Beckworth and Cardiff Garcia

    BTW this is a response to your post on Chinese housing from the same source

    http://synthenomics.blogspot.com/2012/09/chinese-housing-reply.html

    I pretty much agree with everything he says on the subject.

  63. Gravatar of Georges Georges
    27. March 2013 at 08:15

    Geoff,

    your statement is quite strong:

    “I’ve shown how Dr. Sumner is wrong in that post”

    We must not have the same definition of proof ….

    “The stock market is not reacting instantaneously to Fed announcements. Your assumption is wrong.”

    hein? Are you checking the stock market prices? Read Woodford’s Jackson hole paper on this, it should be an eye opener.

    Anyway, I feel there is no way we will advance on that front, clearly you do not want to accept a different interpretations of the facts besides your own, maybe I’m wrong but you have clearly not shown that at all!

  64. Gravatar of libertaer libertaer
    27. March 2013 at 08:19

    Scott,
    I’ve a question which is related to Rob Rawlings’. Imagine an economy with a fixed pool of investment goods. People can only produce consumption goods. Nevertheless, people want to save more (which is impossible since there are no more investment goods), so they hoard the medium of exchange misusing its “store of value”-property and create a fall in AD.

    What do you do now? Open market operations, printing money and buying some of the fixed pool of investment goods, won’t work, because for every investment good people lose, they will hoard money as a substitute. Each open market operations will raise the demand for money by reducing the supply of investment goods.

    Isn’t it right to say that in this case only helicopter drops can work, not open market operations?

  65. Gravatar of Geoff Geoff
    27. March 2013 at 08:25

    Georges:

    “Geoff, your statement is quite strong:”

    “I’ve shown how Dr. Sumner is wrong in that post”

    It was no more strong than Dr. Sumner’s statement that I am wrong.

    “We must not have the same definition of proof ….”

    You don’t know my definition.

    “The stock market is not reacting instantaneously to Fed announcements. Your assumption is wrong.”

    “hein? Are you checking the stock market prices? Read Woodford’s Jackson hole paper on this, it should be an eye opener.”

    Yes, I am checking stock market prices. Yes, I have read that paper. I still hold that stock prices do not instantly adjust to inflation.

    “Anyway, I feel there is no way we will advance on that front, clearly you do not want to accept a different interpretations of the facts besides your own, maybe I’m wrong but you have clearly not shown that at all!”

    I have shown how you are wrong about inflation and whether or not it changes spending and prices at the primary dealer stage. I haven’t shown how you are wrong in the other areas, because I only have two hands.

  66. Gravatar of Geoff Geoff
    27. March 2013 at 08:30

    Georges:

    For the stock price adjustment issue, if you accept that today’s saving and nominal demand for stocks, by you and me and everyone else who earns an income, can affect stock prices today, and if you accept that your income does not instantly rise when Bernanke inflates, but rather than time must pass before such inflation percolates throughout the economy, such that a few months or a year must elapse before you get a raise from past inflation, it follows that you must hold the position that stock prices do not instantly adjust from initial inflation, but are affected over a time period that spans however long it takes for your income to rise due to Bernanke’s inflation into the banking system.

    There are so many people on this blog who are so confused about how inflation works. It’s like there is this widespread denial that inflation affects spending and prices non-homogeneously, that everyone’s incomes are raised at exactly the same rate and at the exact same time, when Bernanke increases the supply of bank reserves.

  67. Gravatar of Georges Georges
    27. March 2013 at 08:45

    Geoff, there have been a lot of work in inflation mechanism and very few fit your narrow criteria.

    What you are talking about stock prices is not correct, if you believe in EMH you would know that what you are saying is wrong, but you don’t even need to believe in EMH, it’s enough to just believe in a very weak EMH version, or in Rational Expectations to see that what you are saying is not exact.

  68. Gravatar of Rob Rawlings Rob Rawlings
    27. March 2013 at 08:56

    “You are thinking in Keynesian terms, and you CANNOT understand monetary economics while thinking in Keynesian terms. It’s not about spending on real goods and services, it’s about the supply and demand for the monetary base”

    I think this is an unfair statement. In all my comments I have been trying to make the same point – that your example lacks all the relevant information needed to see how the change you describe to the monetary base will affect the supply and demand for it – leaving the impression that just changing the composition of someone’s income between cash and check will somehow kick off a hot potato effect.

    I don’t see how this can be described as “Keynesian thinking”

  69. Gravatar of Geoff Geoff
    27. March 2013 at 09:19

    Georges:

    “Geoff, there have been a lot of work in inflation mechanism and very few fit your narrow criteria.”

    There has been a lot of work on the mechanism that fits my criteria, and regardless of the relative size of this work, it has no bearing on the quality or truth of it. In 1900, relativistic physics research was far outweighed by Newtonian physics research. But that didn’t mean relativistic physics was less right.

    You’re teetering very close to ad populum fallacy.

    “What you are talking about stock prices is not correct, if you believe in EMH you would know that what you are saying is wrong, but you don’t even need to believe in EMH, it’s enough to just believe in a very weak EMH version, or in Rational Expectations to see that what you are saying is not exact.”

    EMH is one theory among many that are consistent with historical data.

    I don’t personally adhere to it, because it suffers from numerous internal logic problems.

    Rational expectations is definitely false, because it presumes that every individual has the same knowledge set, and the same expectations set. In the real world, there are disagreements.

    What you are saying is not correct if you accept a real world, economically sound theory.

  70. Gravatar of Georges Georges
    27. March 2013 at 10:09

    Geoff, and what would that real world economically sound theory? yours presumably? :)

    You don’t see how you are going in circles, I’m right hence I’m right. You discard theories off hand because they don’t fit your criteria, and you say if someone does not agree with you then they have not understood the inflation mechanism.

    I’m not saying that what I’m saying or what ssumner is saying is the absolute truth, but I have given you different mechanism where what I’m saying can be true, and you just discard them off hand because they don’t fit your definition of how things work!

  71. Gravatar of Geoff Geoff
    27. March 2013 at 10:34

    Georges:

    “Geoff, and what would that real world economically sound theory? yours presumably?”

    Anyone can learn it and accept it. It isn’t “mine.”

    “You don’t see how you are going in circles, I’m right hence I’m right.”

    I am not making that argument, nor implying it. Can you be more specific?

    “You discard theories off hand because they don’t fit your criteria”

    I discard theories when I notice internal flaws in them.

    “and you say if someone does not agree with you then they have not understood the inflation mechanism.”

    It’s not just about disagreement. It’s about the erroneous positive claims I see you making.

    “I’m not saying that what I’m saying or what ssumner is saying is the absolute truth, but I have given you different mechanism where what I’m saying can be true, and you just discard them off hand because they don’t fit your definition of how things work!”

    It’s not “off hand”. Did you actually believe that your claims are the first time I have ever heard them? I’ve seen them and studied them in detail.

    Perhaps you have accepted your theories without much thought, and so you believe that they can only be discarded the same way? I know you didn’t think of them yourself. You were taught them and you accepted them on the basis of trusting what you perceive to be authorities and your superiors. I question authority, and I have no qualms with rejecting and accepting theories from authorities based on my own accumulated knowledge.

    You still haven’t responded to my challenges. You haven’t even engaged them or recognized them as being challenges.

    I challenged your theory by making the case that it is logically impossible for the Fed to affect aggregate spending and prices, through engaging in OMOs with the primary dealers only, while at the same time not affecting spending or prices at the primary dealer stage.

    You have accepted the claim that the Fed can increase aggregate spending and prices over time through OMOs, but without affecting spending or prices at the primary dealer stage, despite the fact that the primary dealers are (under offical, “normal” circumstances) the only institutions that connect the Fed with the aggregate market.

    You have only evaded this and mentioned other papers, authors and so on, who disagree. Contrary to me rejecting what you are saying on the sole basis that you disagree with me, what you are doing is dismissing what I am saying because you perceive me as disagreeing with you and those of whose theories you have seemingly accepted without much critical thought.

    It is not a flaw in my character to reject your claims because I think they’re wrong.

  72. Gravatar of Max Max
    27. March 2013 at 10:39

    “Max, How can seignorage on currency be negative?”

    If the Fed lowered interest on reserves below 0%, then the Fed Funds rate would go negative.

    The “zero bound” isn’t actually 0%, it’s below 0%.

  73. Gravatar of Tyler Joyner Tyler Joyner
    27. March 2013 at 11:02

    Geoff,

    You said, “There are so many people on this blog who are so confused about how inflation works. It’s like there is this widespread denial that inflation affects spending and prices non-homogeneously, that everyone’s incomes are raised at exactly the same rate and at the exact same time, when Bernanke increases the supply of bank reserves.”

    I think you’re correct on this, but don’t hold your breath waiting for any market monetarists to address it. A lot rests on the idea that the Fed is effective and its operations under NGDPLT will have little or no negative unintended consequences.

  74. Gravatar of Geoff Geoff
    27. March 2013 at 11:08

    Tyler Joyner:

    “I think you’re correct on this, but don’t hold your breath waiting for any market monetarists to address it. A lot rests on the idea that the Fed is effective and its operations under NGDPLT will have little or no negative unintended consequences.”

    Funny, the inflationists said the same thing about price index targeting.

    Meet the new God, same as the old God…

    If we ignore relative spending and price changes, then they don’t exist.

  75. Gravatar of Tyler Joyner Tyler Joyner
    27. March 2013 at 11:27

    Geoff,

    I’m interested in the relationship between organizations that exist to solve a particular problem, and the problem itself. Someone must have addressed it before, but it’s a fun thought experiment anyways.

    If there is a group of people who get paid to solve a problem, what happens if the problem is either found to be permanently insoluble or is solved once and for all?

  76. Gravatar of Fed Up Fed Up
    27. March 2013 at 11:31

    “Fed up, I prefer the base, as it is the aggregate actually controlled by the Fed.”

    I believe banks and bank-like entities create MOA and MOE. That means the central bank is not in total control.

    “Changes in the demand for demand deposits matter only to the extent that they impact the base, or whatever the medium of account is.”

    My point is demand deposits are MOA. No need to impact the monetary base (currency plus central bank reserves).

    BTW, what are Cyriot DDs worth? The MOA must have a price that doesn’t change.”

    Still 1 to 1? There may be less of them or less available for circulation.

    There is one difference between currency and demand deposits, currency is not directly “defaultable” while demand deposits are directly “defaultable”.

  77. Gravatar of Geoff Geoff
    27. March 2013 at 11:43

    “If there is a group of people who get paid to solve a problem, what happens if the problem is either found to be permanently insoluble or is solved once and for all?”

    That group will have an incentive to insist that the non-existent problem is real, and if the problem is real, but solved, then the incentive would be to criticize said solution.

    This incentive can only be checked if there exists incentives to insist the non-existent problem is not real, and if real and solved, to promote the solution. These incentives exist only if there is real competition with the first group, meaning there must be a private competititive market. Without the competition, the “bad” incentive goes unchecked.

    The Fed has the incentive to not promote, not finance, and otherwise not sanction, research and theories that call for competition, i.e. abolition of its monopoly. Those they pay have the incentive to insist non-existent problems are real, and to dismiss solutions to problems that would call for their unemployment. For others wedded to inflation, the incentive is to not reverse one’s intellectual devotion. This is why it is imperative that young people are shown the ideas and theories of monetary competition. I went all the way in my education, and not once did my profs mention, let alone teach, monetary competition.

    It’s not surprising that economists are almost universally ignorant about the nature and economics of laissez faire money production. Most were never taught it in school, and there are few employment opportunities that would reward economics research that investigates the optimality of abolishing the monopoly.

  78. Gravatar of Fed Up Fed Up
    27. March 2013 at 11:46

    “Rob, [price] Inflation depends on monetary policy, taxes play almost no role. So the helicopter drop is only a tiny bit more inflationary than an OMO.”

    I disagree with that. Taxes remove currency and/or demand deposits (MOA/MOE) from circulation.

  79. Gravatar of Georges Georges
    27. March 2013 at 12:03

    Geoff, I did respond to your challenge by invoking the expectation mechanism.

    Even better as my previous example, bond trades at 100, the primary dealer does not hold it (to simplify assuming there is only one primary dealer), but you a “regular” citizen hold it. Central bank announces I will buy as many bonds as possible to drive the price until 110. This is public information. You holder of the bond figure at 110 I will sell my bond. you phone the primary dealer and tell him I know the fed will take it from you at 110 so I will sell it for you at 109.99(ignore the monopoly power of the dealer here as in reality there is more than just one), the primary dealer says ok and then sells it to the fed at 110 at a profit of 0.01 while and other bondholders have made 9.99 profit.

    clearly the “winner” here is not the primary dealer but the bond holder, you agree?

    Now in rates perspective, this has lowered the rate (let’s say from 4% to 1%), if a project is expected to give me a return of 3% previously I would have preferred investing in bonds now I would invest in the project this will drive AD up (more investment less savings) which will drive NGDP up, all of this with the primary dealer being richer by a meager 0.01.

    Where is the flaw in this?

  80. Gravatar of Geoff Geoff
    27. March 2013 at 13:07

    Georges:

    “Geoff, I did respond to your challenge by invoking the expectation mechanism.”

    Expectations alone cannot raise my income. There is the requirement of higher incomes of those I engage in exchange with, and higher incomes of those they exchange with, and so on. The start of this process is the central bank.

    The notion that expectations alone can do it, would imply that everyone’s cash balances are currently practically infinite, and that everyone only slightly adjusts their spending (upwards) when the Fed committee speaks. Absurd of course.

    Expectations cannot make me keep spending more and more money each year, unless my income rises more and more each year.

    “Even better as my previous example, bond trades at 100, the primary dealer does not hold it (to simplify assuming there is only one primary dealer), but you a “regular” citizen hold it. Central bank announces I will buy as many bonds as possible to drive the price until 110. This is public information. You holder of the bond figure at 110 I will sell my bond. you phone the primary dealer and tell him I know the fed will take it from you at 110 so I will sell it for you at 109.99(ignore the monopoly power of the dealer here as in reality there is more than just one), the primary dealer says ok and then sells it to the fed at 110 at a profit of 0.01 while and other bondholders have made 9.99 profit.”

    I already addressed this as not related to the issue. You are contradicting the meaning of primary dealer by assuming the primary dealer isn’t selling bonds to the Fed. You can’t challenge my argument by denying the assumptions upon which my argument rests, unless you show how those assumptions are themselves wrong. Is it wrong to assume primary dealers are the ones selling bonds to the Fed? No.

    “Now in rates perspective, this has lowered the rate (let’s say from 4% to 1%), if a project is expected to give me a return of 3% previously I would have preferred investing in bonds now I would invest in the project this will drive AD up (more investment less savings) which will drive NGDP up, all of this with the primary dealer being richer by a meager 0.01.”

    Not relevant to the issue.

    “Where is the flaw in this?”

    You’re not addressing the challenge. That’s the flaw. You are not addressing how it is possible for aggregate spending and prices to rise, without spending and prices rising at the primary dealer stage, despite the fact that the primary dealers are the conduit between the central bank and the rest of the economy.

    You are also not explaining how your own expectations theory is in itself possible. How in the world can expectations alone make spending and prices rise in the aggregate, the way they have for the last 100 years, without actual inflation of the money supply?

  81. Gravatar of Georges Georges
    27. March 2013 at 13:20

    Geoff, how in my example are the primary dealers not selling the bond to the fed? tje primary dealers are market makers they buy and sell they wash the transaction they are intermediaries. the bond holder sells the bond to the primary dealer who in turn sells the bond to the fed. If your argument is the bond holders in the evonomy are benefiting from this then yes indeed. But this is not what we were arguing, we are arguing that although the fed is giving the new money to the primary dealer it’s not him whobis profiting from this.

  82. Gravatar of Geoff Geoff
    27. March 2013 at 13:29

    Georges:

    “Geoff, how in my example are the primary dealers not selling the bond to the fed?”

    When you said:

    “Even better as my previous example, bond trades at 100, the primary dealer does not hold it (to simplify assuming there is only one primary dealer), but you a “regular” citizen hold it.”

    “tje primary dealers are market makers they buy and sell they wash the transaction they are intermediaries. the bond holder sells the bond to the primary dealer who in turn sells the bond to the fed. If your argument is the bond holders in the evonomy are benefiting from this then yes indeed. But this is not what we were arguing, we are arguing that although the fed is giving the new money to the primary dealer it’s not him whobis profiting from this.”

    You’re still not addressing the issue. After what is it, 5 posts now?, you continue to evade the challenge.

    I can only repeat it:

    You are not addressing how it is possible for aggregate spending and prices to rise, without spending and prices rising at the primary dealer stage, despite the fact that the primary dealers are the conduit between the central bank and the rest of the economy.

    You are also not explaining how your own expectations theory is in itself possible. How in the world can expectations alone make spending and prices rise in the aggregate, the way they have for the last 100 years, without actual inflation of the money supply?

  83. Gravatar of Georges Georges
    27. March 2013 at 13:38

    Geoff,

    Are you reading what I wrote? The bond holder did not initially hold the bond but he subsequently purchased it from the holder and sold it to the fed! It’s the primary dealer who is receiving the new money but he had given “old” to the bond holder, so it does not matter who received the new money and where the money is injected.

    For the expectation bit this would need a much bigger post than the above and would need some rational expectation type modelling, if the simpler argument above is taking us that many posts I will not dare go through thisone :) but I would invite you to read woodford’s paper where he models it quite rigorously.

  84. Gravatar of Georges Georges
    27. March 2013 at 13:43

    if you do not believe at all in RE type of models then you would not believe in the expectation that is reasobable

  85. Gravatar of Georges Georges
    27. March 2013 at 13:57

    sorry in my reply iI meant the primary dealer did not hold the bond initially

  86. Gravatar of Geoff Geoff
    27. March 2013 at 14:14

    Georges:

    “Are you reading what I wrote? The bond holder did not initially hold the bond but he subsequently purchased it from the holder and sold it to the fed! It’s the primary dealer who is receiving the new money but he had given “old” to the bond holder, so it does not matter who received the new money and where the money is injected.”

    You’re still not addressing the challenge.

    “For the expectation bit this would need a much bigger post than the above and would need some rational expectation type modelling, if the simpler argument above is taking us that many posts I will not dare go through thisone but I would invite you to read woodford’s paper where he models it quite rigorously.”

    This doesn’t respond to my expectations challenge.

    “if you do not believe at all in RE type of models then you would not believe in the expectation that is reasobable”

    I don’t know what that means.

    “sorry in my reply iI meant the primary dealer did not hold the bond initially”

    Still not a response.

    This shouldn’t be hard. I’ll repeat a fourth time:

    How it is possible for aggregate spending and prices to rise, without spending and prices rising at the primary dealer stage, despite the fact that the primary dealers are the conduit between the central bank and the rest of the economy?

    How can expectations alone make spending and prices rise in the aggregate, the way they have for the last 100 years, without actual inflation of the money supply?

  87. Gravatar of Georges Georges
    27. March 2013 at 14:23

    Geoff, I give up, there is no point arguing at this point, I just gave you the example how prices can increase (the bond price have increased from 100 to 110) without the primary dealers benefiting although they are the ones selling to the fed and they are the one receiving the money. You had even acknowledged that but said that in my example the primary dealers are not selling the bond to the fed which is not true.

    You keep saying that I’m not answering your point although you are the one not answering mine, but anyway enough comments on this post, clearly we are going no where!

    I don’t understand your answer to the expectation channel, if you believe in the rational expectation model and if you believe woodford’s assumption and that he’s not a crappy mathematician then he clearly shows how expected inflation can create inflation, this is not only woodford but this the whole new classical Philips curve. You are free not to believe in those assumptions, but you cannot say that they are not logically consistent.

  88. Gravatar of Geoff Geoff
    27. March 2013 at 14:54

    Georges:

    “Geoff, I give up”

    Fair enough. Maybe in the future you’ll address the challenge.

    “I just gave you the example how prices can increase (the bond price have increased from 100 to 110) without the primary dealers benefiting although they are the ones selling to the fed and they are the one receiving the money.”

    Bond prices increasing due to higher spending in the private market requires lower spending on other things. You’re not addressing the challenge that considers a rise in aggregate spending and prices, not just a rise here and there.

    “You keep saying that I’m not answering your point although you are the one not answering mine, but anyway enough comments on this post, clearly we are going no where!”

    I would like for my challenge to be addresed first, before I address yours.

    “I don’t understand your answer to the expectation channel, if you believe in the rational expectation model and if you believe woodford’s assumption and that he’s not a crappy mathematician then he clearly shows how expected inflation can create inflation”

    This statement violates the QTM, which holds that the amount of aggregate spending that previals in the economy, is a function of the amount of money that exists in the economy.

    “this is not only woodford but this the whole new classical Philips curve. You are free not to believe in those assumptions, but you cannot say that they are not logically consistent.”

    I can, and I do. Sorry.

  89. Gravatar of Georges Georges
    27. March 2013 at 15:06

    Geoff, I thought you agreed with me that OMOs are inflationary so can you please enlighten on how can this would be happening? Since you just said my example does not show that aggregate prices are going up?

    Well I’m sorry to hear that nobel prize winner economists are just idiots, since they cannot even make up a logically consistent theory. Which brings me back to your initial example of physics, I’m pretty sure no one accuses Newton of being logically inconsistent they just review his assumptions, but no one accuses him of inventing illogical statements, sadly it’s clearly not the case in economics…

  90. Gravatar of Geoff Geoff
    27. March 2013 at 15:16

    Georges:

    “Geoff, I thought you agreed with me that OMOs are inflationary so can you please enlighten on how can this would be happening? Since you just said my example does not show that aggregate prices are going up?”

    I guess I should ask you this:

    How can spending and prices keep going up in the aggregate? Explain the sequential processes over time, if you please.

    “Well I’m sorry to hear that nobel prize winner economists are just idiots, since they cannot even make up a logically consistent theory.”

    Kissinger and Obama won Nobel Peace Prizes. Hitler was Time Magazine Man of the Year.

    I am not saying they’re idiots. Just wrong. They can be wrong, you know.

    “Which brings me back to your initial example of physics, I’m pretty sure no one accuses Newton of being logically inconsistent they just review his assumptions, but no one accuses him of inventing illogical statements, sadly it’s clearly not the case in economics…”

    I didn’t say Newton was logically inconsistent. I invoked Newton to explain to you that just because the majority of academics accept one thing, it doesn’t mean it’s right. You suggested it did, by trying to strengthen your argument by deferring to popularity contests.

  91. Gravatar of Georges Georges
    27. March 2013 at 15:31

    I’m refuting your argument of saying their theories are logically inconsistent that is a very big statement to accuse of! Especially in a scientific domain. You can always argue the assumptions, for instance you can think that rational expectations is wrong this is fair enough, but of you do believe in it and some other new classical assumptions than you would be believing in the consequences of the theory.

    I was hoping that economics was a scientific domain so not to be compared with peace Nobel prize or a time magazine cover, I don’t think anyone has at any time accused a Nobel prize physicist of being logically inconsistent although a lot of physical theories have gone out of use.

    When I was referring to other arguments it’s to refute the fact that you only want your point of view to fit the facts, and you are not willing to accept that some other transmission mechanism can exist as well. This is in no way a popularity contest.

    I have one process which does not even involve OMOs, people lose faith in the currency, no one wants to hold that currency velocity skyrockets and you have hyperinflation.

    As for OMOs,same process as before, bond prices go up, ceteris paribu, the nominal quantity of goods in the economy have increased so NGDP have increased. Now if you believes that it will not be ceteris paribu and that when bond prices go up, some other price has to go down (that other price not being the price of money so the inverse of inflation), then you are saying that an OMO is never inflationary. This would again be another assumption, which you may choose to hold and I’m fine with. But I had the impression that you did not hold that opinion.

    The primary dealers are not getting wealthier from OMOs(if we ignore the tiny commission they earn), I think history did show us that OMOs are inflationary although maybe through different mechanisms than just bond prices going up. If you reason away from the ZLB, I think we can confidently say that as long as the fed is not paying IOR then OMOs are decreasing the interest rate at least, and arguing that neither the MB nor the level of interest rate is not inflationary and also not believing in the expectation, what other way can there be for us to see inflation.

    My honest opinion, I think you do believe that OMOs are inflationary, and you do get my example how in OMOs primary dealers are only earning commissions which is very small compared to the price level impact, but you still just refuse to accept it, because you are holding on to your thought that inflation must propagate like a domino effect and that the people who receive the new money has to benefit from this.
    Or it’s just a total misunderstanding and then I don’t understand what you were arguing from prof sumner’s blog.

  92. Gravatar of Lorenzo from Oz Lorenzo from Oz
    27. March 2013 at 16:17

    It is frustrating how people use the commodity boom as an explanation fetish to stop thinking about the Australian example. First, it does not explain how Australia missed out the previous global recession. Second, the commodity boom also complicates monetary policy because it creates a “two-speed” economy, where some States/regions (Western Australia, Queensland notably) have booming commodity exports and other States/regions (Victoria notably) suffer from the high $A making their services and manufacturing exports more expensive.

  93. Gravatar of Geoff Geoff
    27. March 2013 at 16:33

    Georges:

    “I’m refuting your argument of saying their theories are logically inconsistent that is a very big statement to accuse of!”

    How are you refuting it? I see you denying it.

    “I was hoping that economics was a scientific domain so not to be compared with peace Nobel prize or a time magazine cover, I don’t think anyone has at any time accused a Nobel prize physicist of being logically inconsistent although a lot of physical theories have gone out of use.”

    I was referring to Nobel Peace prize winners so as to show you that winning awards does not necessarily mean one is infallible or irrational or immoral.

    “When I was referring to other arguments it’s to refute the fact that you only want your point of view to fit the facts, and you are not willing to accept that some other transmission mechanism can exist as well. This is in no way a popularity contest.”

    No, that’s what you are doing. I am making an argument that is based on the effects of “spill over”, whereas you are only proposing hypotheticals of how you believe inflation (defined as rising prices) might occur otherwise.

    Again, you’re not challenging the point I am making, but you are instead relegating yourself to making tengential points.

    Regarding expectations, are you saying that expectations alone are sufficient to increasing aggregate spending and prices over time? That central banks don’t have to engage in any OMOs, but just promise that price inflation will be such and such, or that NGDP will be such and such?

    If you accept the obvious fact that something more than mere expectations are necessary in order for aggregate spending and prices to increase over time, then you realize that THAT is the part of inflation that I am referring to in my challenge.

    “I have one process which does not even involve OMOs, people lose faith in the currency, no one wants to hold that currency velocity skyrockets and you have hyperinflation.”

    Why would people lose faith in a currency that is not, according to your thought experiment, being inflated?

    “As for OMOs,same process as before, bond prices go up, ceteris paribu, the nominal quantity of goods in the economy have increased so NGDP have increased.”

    You’re skipping steps in there. How does MY income rise by virtue of the Fed engaging in OMOs? Suppose the Fed buys a bond from a bank. How would my income go up?

    “Now if you believes that it will not be ceteris paribu and that when bond prices go up, some other price has to go down (that other price not being the price of money so the inverse of inflation), then you are saying that an OMO is never inflationary. This would again be another assumption, which you may choose to hold and I’m fine with. But I had the impression that you did not hold that opinion.”

    I’m not saying that.

    “The primary dealers are not getting wealthier from OMOs(if we ignore the tiny commission they earn), I think history did show us that OMOs are inflationary although maybe through different mechanisms than just bond prices going up.”

    The primary dealers are getting wealthier by OMOs, or else they would not sell bonds to the Fed, and not be on the primary dealer list. Obviously their actions are telling you that they are gaining from it.

    “My honest opinion, I think you do believe that OMOs are inflationary, and you do get my example how in OMOs primary dealers are only earning commissions which is very small compared to the price level impact, but you still just refuse to accept it, because you are holding on to your thought that inflation must propagate like a domino effect and that the people who receive the new money has to benefit from this.”

    I do not deny that primary dealers earn commissions. You seem to be denying that there is no domino effect with inflation. But then you have to explain how my income goes up, or how someone else’s income goes up, or, in general, how everyone not the primary dealers’ income goes up, when Bernanke buys bonds from the primary dealers.

    “Or it’s just a total misunderstanding and then I don’t understand what you were arguing from prof sumner’s blog.”

    Again, the challenge I laid out remains. Expectations are not, as shown, sufficient. I think you are just denying the domino effect because you don’t want to accept an uncomfortable truth.

  94. Gravatar of Georges Georges
    27. March 2013 at 23:04

    Geoff, is it not obvious how your income went up and not the primary dealers?? The primary dealer is earning 0.01 commission you are made 10 wealthier by holding the bond! Your income goes up simply because the assets you hold are going up in value!

  95. Gravatar of Geoff Geoff
    28. March 2013 at 05:12

    Georges:

    “Geoff, is it not obvious how your income went up and not the primary dealers?? The primary dealer is earning 0.01 commission you are made 10 wealthier by holding the bond! Your income goes up simply because the assets you hold are going up in value!”

    This again violates the assumption upon which my challenge rests. My challenge assumes there is a rise in aggregate spending and prices, not just a rise in the prices of bonds such that the sellers earn more money. If the income of a bond seller goes up, that requires someone else in the private sector to reduce their cash balance, to reduce their spending on other things such that they have the funds available to purchase the bonds.

    Again, I am asking you how can AGGREGATE spending and prices go up in the private sector over time.

    Don’t assume I am a bond seller. Imagine me to be in the private sector as a wage earner. Not everyone can be a bond seller after all. It is necessary for bond sellers to trade with bond buyers. So imagine how the income of the bond buyers goes up over time due to inflation.

  96. Gravatar of ssumner ssumner
    28. March 2013 at 06:54

    Rob, I have no idea what you are trying to say, can you be more specific.

    Peter, I am assuming zero interest on reserves, which makes reserves equivalent to cash.

    libertaer, Won’t the price of investment goods (or the financial assets that are claims to investment goods) adjust until supply equals demand?

    Rob, You said;

    “that your example lacks all the relevant information needed to see how the change you describe to the monetary base will affect the supply and demand for it”

    This makes no sense. The change in the monetary base IS THE CHANGE IN SUPPLY. And I’m assuming no change in real demand in the long run. I’ve been very specific.

    Max, OK, but I’m assuming no IOR here.

    Fed up, It doesn’t matter what you think, they can’t create base money, and I’m modeling base money here. You need to discuss what I am doing here, not what you wish I was doing.

    You said;

    “My point is demand deposits are MOA. No need to impact the monetary base (currency plus central bank reserves).”

    This is completely wrong. If you have a model of the base, then anything that affects its value must affect the supply or demand for base money. That’s not a theory, it’s a tautology.

    And no, taxes DO NOT remove currency from circulation. Only the Fed can do that.

  97. Gravatar of Rob Rawlings Rob Rawlings
    28. March 2013 at 07:31

    Scott,

    Thanks for your patience – I’m probably a) missing something and b) explaining things badly.

    Let me list my thinking step-by-step and you can tell me where I go wrong.

    1. Some individuals are receiving regular income from the govt in the form of checks

    2. The govt print up a pile of new money. This represent an increase in the monetary base.

    3. The govt mails out all the newly printed money and reduces the value of the checks mailed to those individuals.

    4. If you define the monetary base as bank reserves + paper money then at this point (immediately after the checks have been mailed) the individuals are holding exactly the same amount of money as in previous periods. The increase to the base is all held in the govt acct from which the checks are paid and which will have a bigger balance because the checks were smaller.

    5. The recipients of the checks + new money spend the money on the same things as in prior periods.

    6. If nothing else happens then there will be no hot potato effect and prices will not rise.

    7. If he govt agency that holds the bank acct with the additional balances from the money it saved from mailing checks of a smaller value decides to spend some of that money then this is when the hot potato effects kicks in and leads to an increase in prices.

    8. If the govt agency didn’t spend more then it simply holds larger cash balances. M and V increase and P stays the same.

    (I think the key point may be 6. If you disagree – then why ?)

  98. Gravatar of Rob Rawlings Rob Rawlings
    28. March 2013 at 07:38

    I meant (point 8) M increases and V decreases

  99. Gravatar of Rob Rawlings Rob Rawlings
    28. March 2013 at 07:56

    Perhaps you are assuming that paper money is not interchangeable with money held in banks ? If so then your model makes logical sense : people could only reduce their paper money holdings by spending it – but that seems an odd assumption to make.

  100. Gravatar of Georges Georges
    28. March 2013 at 08:03

    Geoff, OK so let me understand your point, you are not arguing at all the fact that the person who receives the money first is benefiting right? You agree that the primary dealer here is not benefiting at all (well tiny benefit due to commission), but what you are arguing is that the inflation that is happening is happening first in the bond market, and then it’s spilling over the economy vs a one off generalized inflation?

    If this is your point, then this – I think – is very different from your initial point and is not at all contradictory with what ssumner had said:

    “Some Austrians worry about “Cantillon effects,” which means they think it’s important to consider who gets the money first. (Although the term also has other meanings). They assume that that lucky group will boost its spending. Yet the money is not given away, it’s sold at market prices. So the person getting the money first is not significantly better off, and hence has little incentive to buy more real goods and services.”

    Right?

    The discussion on then the mechanism of inflation and if it’s all of a sudden generalized or not is a different one, and I believe personally that you can have both forms.

  101. Gravatar of Geoff Geoff
    28. March 2013 at 09:00

    Georges:

    “you are not arguing at all the fact that the person who receives the money first is benefiting right?”

    I wasn’t arguing that, no, but it is true.

    “You agree that the primary dealer here is not benefiting at all (well tiny benefit due to commission), but what you are arguing is that the inflation that is happening is happening first in the bond market, and then it’s spilling over the economy vs a one off generalized inflation?”

    Are you saying you now agree that there is a spill-over effect taking place with inflation?

    If you concede that, then we’re essentially done, because my challenge is grounded on that point and the other points already addressed. Since you finally admitted it, the next step is to go waaaaay back to where I initially made this point, and continue on:

    If you admit that there is a spill over effect with inflation, then it follows that because real supply doesn’t instantly arise underneath the money flows, it follows that prices are therefore affected non-homogeneously as well. Both spending and prices are changed at the given initial injection points, as compared to no inflation at all, or at different injection points.

    Thus the Cantillon Effect, *properly understood*, is reaffirmed. All the Cantillon Effect holds is that spending and prices are affected heterogeneously with monetary inflation. The spill over effect results in some goods increasing in price before other goods, as the spill over effect increases nominal demand for individual goods in subsequent steps. Since not everyone can be a seller of initially affected goods, it it is logically necessary that only some people benefit from the inflation, whereas everyone else are affected by inflation through paying higher prices. Once the money keeps flowing throughout the economy, then it raises more and more people’s incomes, but those whose incomes are raised later on, have already had to live with paying higher prices in the past. Once their income rises, there is already another round of inflation in the pipeline, again behaving with the spill over effect you now admit takes place.

    “If this is your point, then this – I think – is very different from your initial point and is not at all contradictory with what ssumner had said:”

    “Some Austrians worry about “Cantillon effects,” which means they think it’s important to consider who gets the money first. (Although the term also has other meanings). They assume that that lucky group will boost its spending. Yet the money is not given away, it’s sold at market prices. So the person getting the money first is not significantly better off, and hence has little incentive to buy more real goods and services.”

    Actually it is contradictory, because it does matter who gets the new money first. It matter because there has to be those who pay the higher prices for whatever goods are initially affected, but whose incomes aren’t any higher at that time.

    “The discussion on then the mechanism of inflation and if it’s all of a sudden generalized or not is a different one, and I believe personally that you can have both forms.”

    You cannot possibly have both forms. Sudden general inflation never happens, and cannot happen. Not as long as the central bank does not send everyone checks. As long as they send only some people checks, those people have to spend that money before other people’s incomes can rise, but because that raises prices, those who have to wait for higher incomes, are worse off.

  102. Gravatar of Georges Georges
    28. March 2013 at 09:35

    Geoff,

    Let’s not argue about the fact that you can have a generalized inflation or not as this is a very big topic, but I can give you a mechanism where this could happen (yes it involves rational expectations which you clearly do not believe in, so we can agree to disagree on this point), you are a person in the economy who does not hold a bond and is not a primary dealer, you hear the fed announcing, so you know that the bond price will go up in value so you know that people will be spending money and you will have this spillover effect, so you anticipate this effect and start it now, you start spending more and saving more, it will change your preferences of spending/consumption instantly! Not to believe in that you would to believe in kind of idealized world of RE which you may argue is not very reasonable, but let’s jyst agree to disagree on this.

    For the point on it matters where the money is injected first, I did not understand your sentence:

    “Actually it is contradictory, because it does matter who gets the new money first. It matter because there has to be those who pay the higher prices for whatever goods are initially affected, but whose incomes aren’t any higher at that time.”

    Let’s go over it step by step:

    1) You agree that OMO causin localized inflation (inflating the bond prices), so NGDP is going up.
    2) You agree that it’s the primary dealer who is receiving the money first.
    3) You agree that the primary dealer is only 0.01 richer
    4) You agree that the bond holder is 0.99 richer

    hence you agree that you had a higher NGDP without caring where the money was injected initially.
    What matters if that the fed announced that it will buy bonds, if it announced that it wants to buy gold, then it would have the gold holders who would have benefited even if it’s buying from the primary dealers.

    All of this price increase is what would one strictly define as fiscal policy. Monetary policy per say (in a stricty interpretation) is just the fact that the monetary base has increased. And the way this can impact general prices, I see it as through the expectation channnel.

    Geoff, if you believe in all what I have said above but you do not believe in the expectation channel, I think you would be quite close to a keynesian thinking, where only a move in the short term interest rate matters (so equivalently the bond price changing). Not the new keynesians since they would believe in the expectation channel, but kind of a more “traditional” keynesian, no?

  103. Gravatar of Tyler Joyner Tyler Joyner
    28. March 2013 at 09:54

    Georges,

    Geoff certainly does not need any help in this debate, but a couple of points here:

    1) Your hypothetical scenario illustrates exactly why inflation is not an instantly propagating phenomenon. In your scenario, people who see the Fed announcement first can run out and spend their free cash before prices go up. So the guy down the street who runs a corner store sells his goods at a lower price than they will be tomorrow, because he didn’t know about the Fed announcement yet. People with the proper education and opportunity can successfully predict Fed action (you might find some of them working for, say, a primary dealer), which puts them ahead of the people who are glued to the TV but who have no special predictive abilities.

    2) Not everyone has free cash to spend. Many Americans live from paycheck to paycheck, and regardless of what their expectations for the future may be, they can’t spend any more money until they start making more. In the government, which is an institution massive enough to plan for such things, inflation-based raises happen at most once a year. How long do you think it takes for other businesses to raise wage levels to compensate for inflation?

  104. Gravatar of Georges Georges
    28. March 2013 at 10:09

    Tyler Joyner, you are giving me the arguments against why rational expectation may not be correct, this is fine which is I’m saying we should not go through this discussion here.

    If you believe strictly in rational expectation, everyone will be rational, or at least the people who are not rational will make random “errors” with a mean of 0, so that on the aggregate it behaves as if everyone had rational expectations.

    Your second argument is again a very favorite Keynesian argument of the cash strapped people and why you should be doing more active fiscal policy.

    I’m not saying this is wrong, this is just a different point of view. My point of this argument is to refute the wrong argument of that it matters where money is injected first.

  105. Gravatar of Tyler Joyner Tyler Joyner
    28. March 2013 at 11:53

    Okay, so allow me to focus on your stated point, “to refute the wrong argument of that it matters where money is injected first”.

    Rational expectations is clearly central to your argument, yet you decline to defend it here, and RE is neither universally accepted nor accepted among the particular participants of the discussion. In my view even if the “errors” have a mean of 0, which is an uncertainty in itself, they are not random. Some actors and classes of actors are far more likely to have a positive error, and others are far more likely to have a negative error. Primary dealers fall into the first category. That’s kind of the point.

    Geoff mentioned more than once that primary dealers are benefiting from OMOs, as illustrated by the fact that they become primary dealers of their own free will, a status which is not lightly given nor free from cost. Clearly those institutions disagree with your dismissal of the benefits of being a primary dealer as unimportant.

    I find it difficult to believe that the Fed is buying $85 billion in debt each month, but the people they buy the debt from are not benefiting. Anyone who has ever sold anything will tell you that they would like a guaranteed $85 billion in extra business each month.

    Let’s imagine a city, and call it Atlantis. Atlantis is similar to most western cities in most respects. There is a certain demand for construction companies in the city, from private developers and so forth. Then the Atlantis government announces it will build a new one thousand screen megaplex theater, which will cost $500 million dollars, and the contract has been awarded to the premier construction company in Atlantis, Golden Socks. Now you could argue that the total demand for construction in the city has gone up, and in that sense other construction firms might benefit by getting business that Golden Socks is now too busy to go after properly. However, I find it hard to believe that you think Golden Socks is not benefiting any more than the other construction firms. And yet that seems to be your position.

    “But bonds are financial assets!” you say, “It’s simply one obligation being traded for another! The Atlantis example is fiscal policy!” Glad you brought it up. The US Treasury issues bonds, which are bought by primary dealers and resold later. The Fed buys bonds and remits the interest back to the Treasury. Primary dealers make money on both sides of the trade. You cannot cleanly separate fiscal and monetary policy when the monetary authorities have proven themselves vulnerable to political pressures on several occasions, and when the fiscal authorities are direct beneficiaries of monetary policy. As Geoff said, there are winners and losers.

    If my second argument seemed Keynesian to you, it was not by design. I did not mean to imply that I support running large deficits as a positive alternative to monetary policy. I’m not even making a “End the Fed” argument either, nor attempting to pass judgment on any particular policy. My point is simply that monetary policy has unintended consequences whose existence are not acknowledged, and there are very obvious reasons why acknowledging those effects would be painful for certain people.

  106. Gravatar of Georges Georges
    28. March 2013 at 12:11

    The primary dealers are pass through actors they are financial intermediaries. I thought we had agreed that the primary dealer is only gaining 0.01 and it’s the bondholder who is benefitng the most by earning 9.99 in ny example?

    They are as you sad garanteed 85bn extra business but ther margin is very low on this, and the argument is it’s not this commission that is creating the inflation and from a macro perspective this is peanuts …

  107. Gravatar of Geoff Geoff
    28. March 2013 at 12:17

    Georges:

    “Let’s not argue about the fact that you can have a generalized inflation or not as this is a very big topic, but I can give you a mechanism where this could happen (yes it involves rational expectations which you clearly do not believe in, so we can agree to disagree on this point), you are a person in the economy who does not hold a bond and is not a primary dealer, you hear the fed announcing, so you know that the bond price will go up in value so you know that people will be spending money and you will have this spillover effect, so you anticipate this effect and start it now, you start spending more and saving more, it will change your preferences of spending/consumption instantly!”

    You are assuming I have the cash to spend. You can’t assume that, because my cash balance may be inadequate. After all, if I did try to hoard cash, to plan for this type of thing, then the Fed will just print even more money to offset whatever decline in prices would otherwise result, thus making my cash balance less valuable.

    You seem to be presuming that cash balances are essentially infinite, and that people can simply spend more if they expect more inflation. What about those who can’t spend more until they earn more, because they are investing and consuming out of their entire incomes?

    At any rate, and more importantly, even if every individual did spend more, it doesn’t mean that the spill over effect would no longer exist. It would still exist, as the introduction of new spending *raises prices even further*. Prices are a function of supply and demand. If demand rises, so will prices. This is on top of the increased demand by way of reduced cash balances.

    “Not to believe in that you would to believe in kind of idealized world of RE which you may argue is not very reasonable, but let’s jyst agree to disagree on this.”

    If you are proposing RE, and RE is false, then you’re still stuck at square one.

    “For the point on it matters where the money is injected first, I did not understand your sentence:”

    “”Actually it is contradictory, because it does matter who gets the new money first. It matter because there has to be those who pay the higher prices for whatever goods are initially affected, but whose incomes aren’t any higher at that time.”

    “Let’s go over it step by step:

    “1) You agree that OMO causin localized inflation (inflating the bond prices), so NGDP is going up.”
    “2) You agree that it’s the primary dealer who is receiving the money first.”
    “3) You agree that the primary dealer is only 0.01 richer”
    “4) You agree that the bond holder is 0.99 richer”

    Bold holders cannot get richer unless there are bonds being SOLD at those higher prices. Prices are in the realm of actual exchanges, not hypotheticals. If you say that the price of the bond has increased, then the BUYERS of said bond are LOSING OUT, because they are paying a higher price without their income being raised accordingly.

    Does that finally make sense to you now? You keep divorcing the relevant parties from your analysis. You are not considering everyone you should be considering.

    “hence you agree that you had a higher NGDP without caring where the money was injected initially.”

    Since your reasoning above is incomplete, your conclusion here needs work.

    “What matters if that the fed announced that it will buy bonds, if it announced that it wants to buy gold, then it would have the gold holders who would have benefited even if it’s buying from the primary dealers.”

    Then those who are BUYING gold at the higher price are losing out. For they now have to spend more money to get the same gold.

    See a pattern yet?

    “All of this price increase is what would one strictly define as fiscal policy. Monetary policy per say (in a stricty interpretation) is just the fact that the monetary base has increased. And the way this can impact general prices, I see it as through the expectation channnel.”

    “Geoff, if you believe in all what I have said above but you do not believe in the expectation channel, I think you would be quite close to a keynesian thinking, where only a move in the short term interest rate matters (so equivalently the bond price changing). Not the new keynesians since they would believe in the expectation channel, but kind of a more “traditional” keynesian, no?”

    No.

  108. Gravatar of Max Max
    28. March 2013 at 12:21

    “Max, OK, but I’m assuming no IOR here.”

    I think it would be negative with IOR=0% also, but it’s less certain.

    To the extent that the central bank can control seignorage independently of its target, it should aim for zero, or slightly positive (Milton Friedman’s “optimal quantity of money”). If you have to pay people to take your product, you are producing too much. That’s the situation with the Fed – it is overproducing money.

  109. Gravatar of Geoff Geoff
    28. March 2013 at 12:27

    Georges:

    “The primary dealers are pass through actors they are financial intermediaries. I thought we had agreed that the primary dealer is only gaining 0.01 and it’s the bondholder who is benefitng the most by earning 9.99 in ny example?”

    No, they’re gaining the difference between what bond prices would have been had there been no inflation, and what bond prices are with inflation. This difference cannot be observed because we don’t know what bond prices would have otherwise been.

    There are at least two monetary forces on bond prices, one that raises the prices (liquidity effect) and one that lowers bond prices (inflationary premium effect). There is alwaysthe liquidity effect present which is making the bond prices higher than they would have been had the inflation entered say the consumer goods market directly (Fed buys consumer goods). There is also always an inflationary premium effect, which operates on the side of how much inflation is making consumer goods prices rise.

    The net effect of these two forces may result in bond prices rising over time, falling over time, or remaining unchanged. It cannot be scientifically predicted a priori. It is based on judgments of value, grounded on knowledge and preferences.

    If the Fed inflates, and we observe bond prices falling over time soon after, then the only conclusion one can make here is that it is likely that the inflation premium effect dominated the liquidity effect, such that prices fell, but that the bond sellers gained at the expense of consumer goods sellers, because the liquidity effect benefited the bond sellers and not consumer goods sellers.

    In other words, if the Fed purchased consumer goods instead of bonds, then the inflation premium on bonds might be there, but the liquidity effect would not, since the Fed isn’t buying bonds. So if we see a decline in bond prices then, we can say that the decline in bond prices is more than otherwise would have been had the Fed boughts bonds instead. In other words still, the Fed buying bonds is an implicit subsidy to bond sellers.

    “They are as you sad garanteed 85bn extra business but ther margin is very low on this, and the argument is it’s not this commission that is creating the inflation and from a macro perspective this is peanuts …”

    It’s not just the commission that is important here. It’s the entire supply of new reserves the Fed is creating.

  110. Gravatar of Geoff Geoff
    28. March 2013 at 12:28

    Tyler:

    “I find it difficult to believe that the Fed is buying $85 billion in debt each month, but the people they buy the debt from are not benefiting. Anyone who has ever sold anything will tell you that they would like a guaranteed $85 billion in extra business each month.”

    Bingo.

  111. Gravatar of Georges Georges
    28. March 2013 at 13:47

    I had promised that I would stop before and I did not so promise this is my last comment on this post, anyway I feel we are going no where.

    Let me out it very simply and in as much logical statements that I can formalise it in:

    I’m trying to refute the argument:
    “It matters where the money gets injected first”

    To refute this I imagine the following scenario:
    1) we have one fed, and multiple primary dealers that are in perfect competition with each others (or nearly), none of them hold a bond already.
    2) some people in the economy hold some bonds which are worth 100(so people are willing to buy them at 100).
    3) fed announces: I will be buying bonds at 110 I will be paying for those with newly printed money(or by increasing reserves).
    4) on that announcement, some bond holders thought it’s a good idea to sell bonds now, they call the primary dealer and offer it to him at 110-epsilon, epsilon being the bid-offer spread of the primary dealer (typically in order of 1bp).
    4) primary dealer says OK buys the bond from the bond holder, pays him with his credit card or whatever, then sells this bond to the fed at 110. The fed pays him by increasing his reserve account which he uses to clear his credit card balance.
    5) you agreed that this is inflationary(not saying generalised inflation just inflationary, NGDP will go up).

    So if we assume epsilon << 10, we have created inflation without the primary dealer benefiting anything meaningful (if you also add perfect competition he earned nothing he just earn his margin which is equal to his opportunity cost).

    What mattered here is not the fact that the primary dealer made a profit, but the fact that the monetary base has increased (or you can also view it as the interest rate is going down or whatever other equivalent way but certainly not the fact that the primary dealer earned a commission). So in other words, let us assume the primary dealer decided. It to take any commission on the sale because he thought he can do this for free for the client, this would not be inflationary??

    So as far as I'm concerned I have refuted the hypothesis above.

    You can even see the scenario above in a different way, the fiscal authority wants to build a bridge, one scenario it raises debt and pay for it, then the fed buys that bond. Second scenario, the fed prints some money give it to the fiscal authority who pays the bridge with it. Clearly it's the exact same outcome, in the first scenarios the first person to receive the new money is the bond holder, in the second scenario it's the person who built the bridge, are you telling the 2 scenarios will have different outcomes (if we assume that commissions on bond sale are very low)?

    On that point, I conclude my comments! Sorry for the spam!

  112. Gravatar of Georges Georges
    28. March 2013 at 13:53

    Sorry you should read the sentence :

    Let us assume the primary dealer decided NOT to take any commission on the sale …

    Typing from a mobile which explains the multiple typos …

  113. Gravatar of Tyler Joyner Tyler Joyner
    28. March 2013 at 15:21

    “None of them hold a bond already”

    http://www.newyorkfed.org/research/staff_reports/sr299.pdf

    From the abstract:

    “Using data on U.S. Treasury dealer positions from 1990 to 2006, we find evidence of a significant role for dealers in the intertemporal intermediation of new Treasury security supply. Dealers regularly take into inventory a large share of Treasury issuance so that dealer positions increase during auction weeks. These inventory increases are only partially offset in adjacent weeks and are not significantly hedged with futures. Dealers seem to be compensated for the risk associated with these inventory changes by means of
    price appreciation in the subsequent week.”

  114. Gravatar of Georges Georges
    28. March 2013 at 15:32

    Ahh breaking my promise but Tyler this is irrelevant, in your example bond holders are profiting because they are bond holders and not because they are receiving money first so still refuting the argument.

  115. Gravatar of Tyler Joyner Tyler Joyner
    28. March 2013 at 15:37

    They’re benefiting because they are guaranteed the lowest available price on each new issuance, whereas other bond holders are not. Where did you think the “price appreciation in the subsequent week” came from?

  116. Gravatar of Georges Georges
    28. March 2013 at 15:50

    Again irrelevant, they are benefiting from distorted information, they most likely have much more insider information than all the others, but this is next to the point. The problem here is the inefficiencies in the primary dealer market, and not the fact that they are receiving money first.

    So formulating it again, primary dealer are most likely benefiting from some private information that is making them front play the market, but it’s not the fact that they are doing that that is creating inflation. Do you agree that if they behaved in a super optimal way or if they benefited from private information and some other fractions, this will not have a major impact on inflation (except if the market is super distorted)?

    Btw how can Austrians believe in so many market failures and still believe that market clear and that the best thing is to leave unfettered markets? Isn’t this a recipe for disaster? Anyway this is not relevant to the point

  117. Gravatar of Geoff Geoff
    28. March 2013 at 16:02

    Georges:

    “I’m trying to refute the argument: “It matters where the money gets injected first””

    “To refute this I imagine the following scenario:”
    “1) we have one fed, and multiple primary dealers that are in perfect competition with each others (or nearly), none of them hold a bond already.”
    “2) some people in the economy hold some bonds which are worth 100(so people are willing to buy them at 100).”
    “3) fed announces: I will be buying bonds at 110 I will be paying for those with newly printed money(or by increasing reserves).”
    “4) on that announcement, some bond holders thought it’s a good idea to sell bonds now, they call the primary dealer and offer it to him at 110-epsilon, epsilon being the bid-offer spread of the primary dealer (typically in order of 1bp).”
    “4) primary dealer says OK buys the bond from the bond holder, pays him with his credit card or whatever, then sells this bond to the fed at 110. The fed pays him by increasing his reserve account which he uses to clear his credit card balance.”
    “5) you agreed that this is inflationary(not saying generalised inflation just inflationary, NGDP will go up).”

    The primary dealers benefit in your scenario by the difference between the value of bonds had the Fed inflated the money supply to purchase consumer goods, versus the value of bonds had the Fed inflated the money supply to purchase bonds.

    If the Fed announces it will buy bonds at 110, then the gain to the primary dealers is 110 minus the bond price that otherwise would have prevailed had the Fed bought consumer goods.

    The gain to the primary dealers and the loss to the consumer goods companies, is why it matters who gets the new money first.

    “So if we assume epsilon << 10, we have created inflation without the primary dealer benefiting anything meaningful (if you also add perfect competition he earned nothing he just earn his margin which is equal to his opportunity cost)."

    He gains the difference between the Fed buying bonds and the Fed buying consumer goods, or not buying anything at all, in which case the bonds will be priced lower.

    "What mattered here is not the fact that the primary dealer made a profit, but the fact that the monetary base has increased (or you can also view it as the interest rate is going down or whatever other equivalent way but certainly not the fact that the primary dealer earned a commission). So in other words, let us assume the primary dealer decided. It to take any commission on the sale because he thought he can do this for free for the client, this would not be inflationary??"

    The monetary base is not owned by any single party. It is the sum total of all the private accounts plus the government's account. The Fed increases the base by increasing only some people's accounts.

    "So as far as I'm concerned I have refuted the hypothesis above."

    Sorry, but your "refutation" doesn't address the challenge, nor is it even correct in its own right.

  118. Gravatar of Geoff Geoff
    28. March 2013 at 16:03

    Georges:

    “Btw how can Austrians believe in so many market failures and still believe that market clear and that the best thing is to leave unfettered markets? Isn’t this a recipe for disaster? Anyway this is not relevant to the point”

    It isn’t a failure of the market that the government interferes with it!

  119. Gravatar of ssumner ssumner
    29. March 2013 at 04:00

    Rob, You said:

    “If you define the monetary base as bank reserves + paper money then at this point (immediately after the checks have been mailed) the individuals are holding exactly the same amount of money as in previous periods.”

    No they aren’t holding the same amount. Currency is part of the base and checks are not. So the base increases by $200 per recipient. Then the hot potato effect takes over as people try to get rid on non-interesting bearing base money that they don’t want to hold.

    Yes, I assume paper money is not interchangeable with deposits in banks, which is about the most uncontroversial concept I can imagine. Are people indifferent between carrying $40,000 in their wallets, and $40,000 in demand deposits?

  120. Gravatar of Rob Rawlings Rob Rawlings
    29. March 2013 at 07:51

    OK, I think I can see how your example works.

    For me it is easier if one starts with an example where base money consists of gold coins that can either be used as circulating currency or held by private banks as reserves for lending.

    The govt mines and mints 200 extra gold coins per person and sends them out in the same way as before (reducing the value of the checks like in your example).

    In this case I can see how looking at things in terms of demand to hold gold coins this would lead to a decline in the value of gold via a hot potato effect. Substitute “paper money + bank reserves” for gold coins and I think that is your example.

    Is the above correct ?

    If so: Then I still think there is some detail one needs to understand before seeing how the transmission mechanism for the increase in P works. For example: If neither the individuals nor the govt increase their spending then banks will have additional reserves that they will have to lend our or spend on assets for the increase in P to happen. It was this process I was trying to describe above.

  121. Gravatar of Fed Up Fed Up
    29. March 2013 at 12:55

    “Fed up, It doesn’t matter what you think, they can’t create base money, and I’m modeling base money here. You need to discuss what I am doing here, not what you wish I was doing.”

    I’m modeling the real economy. It has currency, central bank reserves, and demand deposits.

    “You said;

    ‘My point is demand deposits are MOA. No need to impact the monetary base (currency plus central bank reserves).’

    This is completely wrong. If you have a model of the base, then anything that affects its value must affect the supply or demand for base money. That’s not a theory, it’s a tautology.”

    If there is a 0% reserve requirement or something equivalent and no desire for extra currency, then the monetary base would not be affected by private demand deposit creation, while these new demand deposits could affect prices. Think mortgages in Canada.

    “And no, taxes DO NOT remove currency from circulation. Only the Fed can do that.”

    By circulation, I mean circulation in the real economy. Putting $20 of currency under a mattress removes it from circulation.

  122. Gravatar of Fed Up Fed Up
    29. March 2013 at 13:29

    “No they aren’t holding the same amount. Currency is part of the base and checks are not. So the base increases by $200 per recipient.”

    That is not the way the accounting people have explained to me. I believe we are assuming this is the gov’t (with emphasis) spending, so let’s use Social Security as an example. Let’s say someone is getting $1,500 a month as a direct deposit. Next, the gov’t adds an extra $200 (unspecified) the next month. For the first case, use direct deposit. The recipient’s checking account gets marked up by $200 (demand deposits) and (with emphasis) the reserve account of the recipient’s bank gets marked up by $200 (central bank reserves) also. This is also happening with $1,500 direct deposit per month. For the second case, use $200 in currency. In the first case, the monetary base went up by $200 of central bank reserves, while in the second case, the monetary base went up by $200 in currency. Back to the second case, the person decides to take the $200 in currency to the bank and deposit it. At the end, it is the same as the direct deposit, the recipient’s checking account gets marked up by $200 (demand deposits) and (with emphasis) the reserve account of the recipient’s bank gets marked up by $200 (central bank reserves) also.

  123. Gravatar of Fed Up Fed Up
    29. March 2013 at 13:32

    EDIT: “have explained to me”

    TO: “have explained it to me”

  124. Gravatar of Peter N Peter N
    29. March 2013 at 17:22

    Let’s go step by step.

    Depending on preference the customer takes cash to the bank and deposits it or keeps the cash. Alternatively the customer takes the check to the bank, deposits it and withdraws cash or not depending on preference.

    At this point the results are identical for the customer – cash equaling cash preference.

    Assuming the customer doesn’t withdraw any money. The bank either ends up with $200 extra cash or $200 in its reserve account. With the reserve the bank can either invest it, ask for the money in cash from the Fed or lend it to another bank at the federal funds rate.

    With the cash, the bank will either keep it if it needs more vault cash and invest $200 of its reserves, send the $200 to the fed and invest the resulting reserves or sell it to another bank for a small fee (small banks deal in cash with the fed through larger bank intermediaries).

    How about a thought experiment? Suppose the government decides to pay $200 a month of everyone’s social security payments in cash (adjusted for inflation) forever. What effect does that have?

  125. Gravatar of Fed Up Fed Up
    29. March 2013 at 17:35

    Rob Rawlings said: “Perhaps you are assuming that paper money is not interchangeable with money held in banks ? If so then your model makes logical sense : people could only reduce their paper money holdings by spending it – but that seems an odd assumption to make.”

    ssumner said: “Yes, I assume paper money is not interchangeable with deposits in banks, which is about the most uncontroversial concept I can imagine.”

    I’d call it controversial. When I say currency and demand deposits are 1 to 1 convertible, I mean both ways.

    “Are people indifferent between carrying $40,000 in their wallets, and $40,000 in demand deposits?”

    Depends on what you mean by indifferent. As long as there is deposit insurance or a bank with extremely low risk assets, I can count on 1 to 1 convertibility both ways. That means both have the same purchasing power. Most people will take demand deposits because currency can be stolen, lost, or destroyed (fire, eaten by a dog, or whatever else).

    Saving $40,000 in currency or saving $40,000 in demand deposits means that I don’t consume, I don’t invest (NIPA definition), or I don’t financially invest in a financial asset that can go down in value.

  126. Gravatar of Fed Up Fed Up
    29. March 2013 at 17:51

    Peter N said: “Depending on preference the customer takes cash to the bank and deposits it or keeps the cash.”

    So are you saying paper money (cash or currency) is interchangeable with deposits in banks?

  127. Gravatar of Peter N Peter N
    29. March 2013 at 19:25

    I’m saying that both the bank and the customer can end up with identical assets from either scenario, as long as the Fed is willing to let banks exchange reserves for cash and cash for reserves at will.

    There has to be some difference, because some people don’t have bank account. However, as of this month, the government is paying most benefits by direct deposit.

    Try the thought experiment. What if tomorrow the government payed all benefits in cash (we’ll ignore the obvious impracticality of this). If sending $200 in cash once is stimulative, you’d think that paying everyone in cash always would be absolutely explosive. Factory workers used to be payed this way.

    I remember when payroll involved a Brinks truck.

  128. Gravatar of Fed Up Fed Up
    29. March 2013 at 20:38

    “How about a thought experiment? Suppose the government decides to pay $200 a month of everyone’s social security payments in cash (adjusted for inflation) forever. What effect does that have?”

    Are people getting $200 less of demand deposits in their checking account and their bank getting $200 less in central bank reserves?

    “There has to be some difference, because some people don’t have bank account.”

    I assume they use a check cashing service or go somewhere that accepts their SS check as a form of payment (MOA/MOE).

  129. Gravatar of Fed Up Fed Up
    1. April 2013 at 19:34

    ssumner said: “Yes, I assume paper money is not interchangeable with deposits in banks, which is about the most uncontroversial concept I can imagine.”

    So you are telling us that we can’t take $100 in currency (paper money) to say a JP Morgan bank branch, deposit the currency (paper money), and get demand deposits (markup of a checking account)?

  130. Gravatar of Angry Bear » Asset Reflux Disease: Explaining Koo to Krugman Angry Bear » Asset Reflux Disease: Explaining Koo to Krugman
    1. July 2013 at 06:11

    […] is exactly the notion that Scott Sumner bruits here, calling it “the concept that lies at the heart of […]

  131. Gravatar of Asymptosis » Asset Reflux Disease: Explaining Koo to Krugman Asymptosis » Asset Reflux Disease: Explaining Koo to Krugman
    1. July 2013 at 06:29

    […] is exactly the notion that Scott Sumner bruits here, calling it “the concept that lies at the heart of […]

  132. Gravatar of Asset Reflux Disease: Explaining Koo to Krugman | Fifth Estate Asset Reflux Disease: Explaining Koo to Krugman | Fifth Estate
    1. July 2013 at 14:54

    […] is exactly the notion that Scott Sumner bruits here, calling it “the concept that lies at the heart of […]

  133. Gravatar of A Simple Post-Keynesian Alternative to IS-LM | Decisions, Decisions, Decisions A Simple Post-Keynesian Alternative to IS-LM | Decisions, Decisions, Decisions
    26. July 2013 at 03:42

    […] The traditional objection to this derives from the classical dichotomy that money must be neutral as per Scott Sumner […]

  134. Gravatar of Bailey Bailey
    28. September 2015 at 09:22

    Hi,

    I’m Bailey from Mauldin Economics, following up on a collaboration proposal I sent you a while back regarding your article.

    I just wanted to remind you that we’ve published our own in-depth, straightforward explanation of inflation: “What is Inflation? A Mauldin Economics Guide to Inflation, Deflation, and Everything In-Between” (http://www.mauldineconomics.com/resources/a-mauldin-economics-guide-to-inflation-deflation-and-everything-in-bet).

    We would love if you would consider sharing it with your readers, as it may help them get a better grasp of the causes of inflation and possibly increase their engagement with your content.

    Either way, I would be happy if you’d take a look at our article, and like I said before, would love to hear from you if you have any other suggestions on ways we could collaborate.

Leave a Reply