Is it OK to lie for pedagogical reasons?

Most of us macro teachers work with some sort of new Keynesian model.  This is roughly the AS/AD model, with an upward sloping SRAS curve.

We teach the model by repeatedly lying to our students.  I’ll give a couple examples:

1.  We argue that the Keynesian model superseded a “classical model” that had a vertical AS curve, and assumed flexible wages and prices.  In fact, no such classical model existed prior to 1975.  Instead, the old Keynesian model replaced something closer to the modern new Keynesian model.  The model of Fisher/Hawtrey/Cassel and many other interwar economists featured sticky prices, short-run non-neutrality of money, and a self-correcting mechanism that brought us back to the natural rate on the long run.  In Fisher’s case there was even a Phillips Curve.

The textbook story is a good pedagogical device.  But it’s not true.  It’s not even close to being true.

2.  We explain that inflation can be produced by fiscal stimulus, supply shocks, and/or monetary stimulus.  Then we talk about the Great Inflation mostly by referring to two shocks.  The first was the huge fiscal stimulus of the 1960s, when LBJ ran up huge deficits to finance the Vietnam War and the Great Society without raising taxes.  Except that this never happened.  Then in the 1970s OPEC jacked up oil prices twice, and this caused the high inflation of the 1970s and early 1980s as the SRAS curve shifted to the left.  Another lie, as SRAS was shifting to the right during the 1970s.  The folly of the oil shock theory was illustrated in the 2000s when oil soared from $20 to $147, and core inflation barely budged.  People were actually surprised by that fact!

If oil shocks are going to raise inflation, they should lower RGDP.  Yet even during the worst of the oil shock period (mid-1973 to mid-1981) RGDP rose by 2.6% per annum.  So the inflation was almost all monetary, even if you generously assume RGDP growth would have been 3.6% in the absence of the oil shocks.  In fact, growth slowed sharply after 1973 for reasons mostly unrelated to oil; the rapid improvement in products like jet airliners and home appliances came to a screeching halt.  We shifted toward a slower growing service economy.  We couldn’t even average 3.6% growth in the 1990s, when oil got cheap and the computer revolution took off.

Are these lies justified?  It’s nice to give students some real world examples of fiscal shocks and supply shocks.  But what if the message they take from this exercise is that monetary explanations of inflation and NGDP determination are “just a theory,” just as evolution or global warming are “just a theory.”  What if the public doesn’t realize that the Fed drives the nominal economy?  Might that lead to less effective public policies?  Might that have contributed in some small way to the fiasco of 2008?

I think our students can handle the truth.  Why don’t we stop lying?”

PS.  If you are wondering why the LBJ story is a lie, recall that the huge budget deficits began under Reagan, and were associated with a big fall in inflation.


Tags:

 
 
 

39 Responses to “Is it OK to lie for pedagogical reasons?”

  1. Gravatar of Geoff Geoff
    23. February 2013 at 08:52

    Friedrich von Wieser, Eugen von Bohm-Bawerk, Carl Menger, and others during the subjectivist revolution in Europe, modeled the supply curve as vertical (i.e. sellers of goods do not attach direct marginal utility to the units they sell), and they assumed flexible wages and prices (i.e. according to supply and demand).

    This was pre-1975, but they weren’t “classical” models.

    Perhaps the word “classical” was used sloppily by Keynesians to characterize all pre-Keynesian economics, and non-Keynesians adopted that terminology out of convenience.

  2. Gravatar of Geoff Geoff
    23. February 2013 at 09:00

    “The folly of the oil shock theory was illustrated in the 2000s when oil soared from $20 to $147, and core inflation barely budged. People were actually surprised by that fact!”

    Monetary inflation doesn’t affect all goods and services equally. It is entirely possible, indeed it is likely, that the effects of inflation would be concentrated in some sectors and not others.

    It is a great mistake to look at core indexes, and reason from them, by claiming something like “Since core inflation indexes was X% (or not X%), it means we can (or cannot) conclude that monetary inflation was (or was not) primarily responsible for the run up of prices in sectors A, B and C.”

    Inflation doesn’t work that way. Price indexes can rise, fall, or stay the same, and the Fed could be blowing up bubbles in specific sectors as the money flow concentrates in that sector more than others.

    Just think what you would do if you suddenly came into ownership of twice or three times your income. Would you buy two or three times more of everything? Of course not. You would buy more of certain things before others. What’s true for you is true for everyone else. Bankers and investors who find themselves flush with new reserves from the Fed won’t purchase double the mortgages on houses, double the gasoline, double the table salt. No, they would probably speculate more in specific sectors, raising their prices first, before “aggregate prices” rise as much.

  3. Gravatar of Jonathan Finegold Jonathan Finegold
    23. February 2013 at 09:01

    Geoff, regarding Böhm-Bawerk I don’t think this was always true. In his theory of price determination two factors of the marginal pairs are the valuations sellers attach to their goods.

  4. Gravatar of Geoff Geoff
    23. February 2013 at 09:05

    It is entirely possible that oil speculators believed that the Fed would soon loosen monetary policy in 2008, and that this would affect oil more so that other commodities, and so rather than waiting for the money to actually be created and waiting for that money to raise the prices of most consumer goods down the road, the oil speculators may have bid up the price of oil first, stored it, and hoped to make a profit as the price increases down the road from the Fed’s actual inflation.

    It is also possible that once the actual inflation was less than the predicted inflation, then oil prices came back down. Then we could say that the story of the oil price shock was monetary in origin, and looking at past aggregate data may make the historian conclude “Since the index barely moved, inflation therefore had little to do with the oil price shock.”

    Never reason from an index.

  5. Gravatar of ssumner ssumner
    23. February 2013 at 09:16

    Geoff, There are lots of problems with your second post. I have lots of earlier posts pointing out that money doesn’t go “into” markets. In any case, it has no bearing on my claim that higher oil prices don’t cause high inflation, even if you were right.

    The Fed has no ability to “blow up bubbles in specific sectors.” It boosts NGDP. Prices in specific sectors reflect microeconomic factors, or cyclical factors.

    The classical economists believed the LRAS was vertical, as do the new Keynesians.

  6. Gravatar of Geoff Geoff
    23. February 2013 at 09:21

    Jonathan:

    See Bohm-Bawerk, “Capital and Interest”, Sennholz and Huncke translation. Spring Mills, Penn., 1959. Volume 2, pg 244.

  7. Gravatar of Jonathan Finegold Jonathan Finegold
    23. February 2013 at 09:30

    Geoff, that’s exactly the book I have in mind. Look up Böhm-Bawerk’s discussion of the marginal pairs.

  8. Gravatar of david david
    23. February 2013 at 09:39

    Hawtrey/Cassel were both considered heterodox in their time, were they not.

    Qua Glasner:

    .. Hawtrey and Cassel occupied a middle ground between orthodox advocates of a kind of “rules of the game” gold standard, which never really existed before World War I and whose establishment after the war was potentially disastrous, and advocates of managed money such as Keynes and Fisher who opposed restoring anything like the prewar gold standard, and preferred the stabilization of domestic prices levels to the restoration of an international currency.

    The revisionism was mutual – the orthodox advocates also wished to claim that their scheme stemmed from prewar tradition and practices, part of a return to normalcy.

    Continuing:

    Although occupying a middle ground on policy, Hawtrey and Cassel were analytically much closer to Keynes, Fisher and other proponents of price-level stabilization or “managed money” than to more orthodox proponents of the gold standard. Prominent, but by no means alone, among theorists who advocated restoring the gold standard in what they believed was its prewar “rules of the game” form were such French theorists as Charles Rist and Jacques Rueff, the idiosyncratic, hyperclassical English economist, Edwin Cannan, and a group of economists who were then developing an “Austrian” theory of the business cycle synthesizing the capital theory of Bohm-Bawerk, the monetary theory of Wicksell, and the policy orientation of the English Currency School. Most notably these included Ludwig von Mises, Friedrich Hayek, Gottfried Haberler, and Fritz Machlup, as well as a young English theorist, Lionel Robbins. The Austrians (e.g., Hayek, 1932/1984, pp. 119-20) criticized Hawtrey and Cassel for advocating a managed gold standard rather than the orthodox version based on the “rules of the game.”

    The pro-orthodoxy weren’t “classical”, but they did want to claim to be the proper successors to the classical economists, rather than letting the Marshallians (Edgeworth, Viner), take the crown.

    Of course, today Marshall dominates pedagogy rather than the Austrians, which only amplifies the confusion.

  9. Gravatar of Becky Hargrove Becky Hargrove
    23. February 2013 at 09:40

    The money we hold collectively is the optimal capacity we have to spend in any given moment, unless of course a contrary notion prevails to hide that fact – i.e. the great lie. Good post. Now to get at the real root of the desire to stop money printing: fear of disruptive innovation in the very sectors where innovation would make the most difference.

  10. Gravatar of Geoff Geoff
    23. February 2013 at 09:49

    Dr. Sumner:

    “Geoff, There are lots of problems with your second post. I have lots of earlier posts pointing out that money doesn’t go “into” markets. In any case, it has no bearing on my claim that higher oil prices don’t cause high inflation, even if you were right.”

    I didn’t say inflation goes “into” markets either. I was saying newly created money isn’t spent on everything equally. It is “spent” in non-linear ways throughout the economy.

    The Fed could engage in OMOs, and the receivers of said money go out and spend that additional money on particular commodities, rather than “everything”. I said this is likely rather than just possible, because nobody buys a bit of everything when their income rises. Bankers and other initial receivers of Fed money will necessarily increase their spending in particular sectors, concentrating in their particular specialized markets, such as real estate, or energy stocks, or whatever the case may be.

    Imagine your income doubling. Would you double your existing expenditures? Certainly not. You would most likely concentrate your additional spending in only some sectors, and perhaps you might even enter a new market that you didn’t penetrate before. You might even REDUCE some of your expenditures (for “cheap” goods that you no longer have to settle for). In other words, the effect of your spending with increased cash is going to have non-linear effects.

    Are you saying it is IMPOSSIBLE for you to have similar responses to increased cash as others, such that you and others concentrate your spending in a particularly “attractive” sector, which will historically result in inflation “blowing up” those sectors while aggregate indexes “barely move”? Impossible? Not a chance? Zero probability?

    “The Fed has no ability to “blow up bubbles in specific sectors.” It boosts NGDP. Prices in specific sectors reflect microeconomic factors, or cyclical factors.”

    Micro-economic factors INCLUDE money changes. If changes in money affects the individual’s value judgments (and we have seen that they do, then money is not purely a macro-economic factor. It is also a micro-economic one.

    As such, the Fed does not only “boost NGDP.” Really, I mean just because that’s the only statistic you look at, it doesn’t mean it is the only statistic affected.

    Changes in money affects people’s valuations of goods in different ways. It doesn’t just change price levels. It affects relative marginal valuations of goods.

    The Fed does not have to intentionally blow up sector specific bubbles in order for them to actually bring them about.

    The Fed boosts whatever particular markets are affected by an increased quantity of money in the possession of the people who receive the new money, and whose valuations are affected that are something other than “I will purchase an additional amount of everything”, which is not likely, but is required in order for your claim to hold up.

    Not everyone has the same subjective value scales, such that they have the same interest in the same basket of goods. The Fed sending checks to investment bankers first will have vastly different affects on relative prices as compared to sending checks to low income families first.

    And no, it is not true that just because the money will soon be re-deposited into a bank no matter who the Fed sends the money to, that the inflation has similar effects no matter who gets the money first. For even though the money is soon in the physical possession of the banks, there is still a difference in who has the purchasing power. The Fed sending checks to low income families will almost certainly have a larger initial effect on basic consumer goods prices and lower effect on stock prices as compared to sending checks to investment bankers.

    The MM claim that inflation instantly affects the prices of everything is, quite frankly, just absurd. It would require low income families, for example, to increase their spending before their incomes rise in line with the Fed’s inflation, before their incomes rise. It would also require those families to have knowledge of what Bernanke is doing everyday.

    “The classical economists believed the LRAS was vertical, as do the new Keynesians.”

    I thought you said it was a lie that the classicals believed the AS curve was vertical:

    “We argue that the Keynesian model superseded a “classical model” that had a vertical AS curve, and assumed flexible wages and prices. In fact, no such classical model existed prior to 1975.”

  11. Gravatar of Geoff Geoff
    23. February 2013 at 09:52

    Jonathan:

    Not sure if you have the same text as me, but in that reference I am reading Bohm-Bawerk say it is more reasonable to consider the seller as attaching no marginal utility to his supply of goods; that he is willing to accept any price for his goods as determined by the competition among his buyers.

    Does that not imply a vertical supply curve?

  12. Gravatar of Wophugus Wophugus
    23. February 2013 at 10:10

    About ten years ago, when I was in high school, my AP macro teacher taught that the inflation of the 70’s was all about monetary policy and that the oil shock was a red herring. So not everyone lies!

    Incidentally, it probably isn’t a coincidence that my first (and last) formal exposure to macroeconomics was at the hands of a staunch monetarist, and then 10 years later I find myself reading (and mostly agreeing with) this blog.

  13. Gravatar of Jon Jon
    23. February 2013 at 10:25

    Scott, the Reagan deficits weren’t particularly large. They seemed large in nominal terms because inflation had been so high. The story of the Reagan deficits is another case of money illusion.

    http://research.stlouisfed.org/fredgraph.png?g=fQY

    The graph uses constant dollars, population normalized. Primary budget balance.

    Equally large deficits reigned before reagan and are nothing compared to the current deficit.

  14. Gravatar of Ben Ben
    23. February 2013 at 10:50

    Dr. Sumner: I do not dispute your argument that monetary policy was a significant contributor to inflation in the 1970s. However, if the SRAS curve shifted right during the 1970s, why was their stagflation in the 1970s? Unless I am mistaken, stagflation can only be caused by a leftward-shift of the SRAS curve.

  15. Gravatar of Bill Woolsey Bill Woolsey
    23. February 2013 at 10:53

    Consider three statements:

    1. There is some level of prices and wages that will result in a level of real expenditure equal to potential output and so full employment of labor and other resources. In particular, if real expenditures are less than potential output and there is high unemployment of labor, there is some lower level of prices snd wages that will result in real expenditure rising to potential output and a return to full employment.

    2. The level of prices and wages always adjust instantly and smoothly so that the level of real expendtiture always remains equal to potential output and so there is always full employment of labor and other resources.

    3. If there is any imbalance between real expenditure and potential output, the best policy approach is to allow the private sector to adjust prices and wages so that real expenditure adjusts to potential output, allowing for a return to full employment. In particular, the government’s budget should remain balanced and the quantity of money should change dollar-for-dollar with gold inflows and outflows. (Or some other version of the gold standard.)

    I see all of these as being quite distinct.

    Old Keynesians sometimes challenged 1, but used 2 as a strawman. 2 was not a common view before Keynes, though it seems that “new classicals” did adopt it in the seventies. There were plenty of people who favored something like 3 before Keynes. It would be pretty painless if 2 were true, and would seem pretty much absurd if 1 were false. However, if 2 is false and 1 is true, then it just means that sticking to the policy rules would involve a somewhat painful adjustment process in some circumstances.

    Of course, Fisher, Cassel, and Hawtrey tended to go with money reform so that the nominal quantity of money would adjust in a way that would avoid the need to make the adjustments in prices and wages. This is inconsistent with 3, and seems pointless if 2 is true, but is quite consistent with 1 being true.

    I believe 1 is true, 2 is false, and don’t favor the policy approach in 3.

    On final note. If you are a small open economy on a gold standard, then most of monetary economics is pretty much irrelevant. The answer to unemployment of labor is lower nominal and real wages to encourage exports and domestic import competing industry. That domestic workers can buy less is of little importance when most of what they buy is imported and their products are mostly exported. With open capital markets, interest rates, especially long term ones, are fixed by world capital markets. Nominal and real income depends on terms of trade and international capital flows. A large capital inflow and “credit” boom, should be a bit worrisome. If the domestic banks receive large amounts of foreign deposits and lend into some bad investment boom, things will be dicey when those foreign depositors want to withdraw gold. Maybe the central bank should “lean against the wind” and accumulate large gold reserves. It will need them to protect the gold parity in the face of large gold withdrawals later.

    Of course, if there is a world imbalance between the supply and demand for gold, this “small open economy” business all falls apart.

    And if you are the Fed in 1920 and hold a large fraction of the world gold stock, and your economy, and especially finanical markets, are no longer small relative to the rest of the world, “small open economy” thinking can be very misleading.

  16. Gravatar of Don Geddis Don Geddis
    23. February 2013 at 11:32

    @Geoff: Your problem is that you are making assumptions you don’t even realize, which are not shared on this blog (esp. by Scott). So most of what you say is irrelevant, because you’re using the wrong model. I’ll point out your mistakes, but you should read many of Sumner’s earlier postings on the subject, if you want to understand better.

    You write: “The Fed could engage in OMOs, and the receivers of said money go out and spend that additional money on particular commodities

    The “receivers” of Fed OMO purchases, have essentially zero change in their behavior, whether or not the Fed engaged in OMOs. Either way, the receivers used to own some T-bills, and they sold them on the open market, and recieved cash for their bonds. There is no reasonable sense in which those private sellers of bonds, have “more” money because the Fed engaged in the OMO, than they would have if the Fed had not done so. So it’s a complete red herring for you to discuss how the sellers are then going to change their future purchase behavior.

    Imagine your income doubling.

    But that’s not a relevant scenario, because the sellers of the Treasuries don’t find that their income is any (significantly) different, whether or not the Fed does OMOs. So this microeconomic imagining of yours never occurs.

    The Fed boosts whatever particular markets are affected by an increased quantity of money in the possession of the people who receive the new money

    Sumner was trying to hint to you, that you’re engaging in a widespread fallacy that it matters “who gets the new money first”. He has addressed this in the past, and has numerous posts debunking the fallacy. That is simply not the mechanism by which Fed OMOs raise inflation. It is not a flow of cash, from first point of injection, only later to the rest of the economy.

    The Fed sending checks to investment bankers first will have vastly different affects on relative prices as compared to sending checks to low income families first.

    Another irrelevant scenario. The Fed doing OMOs (buying Treasuries on the open market) is a very different operation, from the Fed directly sending “helicopter drops” of new cash to specific individuals. Your new scenario does indeed have distortion effects, in a way that OMOs don’t.

    It would require low income families, for example, to increase their spending before their incomes rise in line with the Fed’s inflation, before their incomes rise.

    Expectations of future inflation (income & prices), do indeed change people’s immediate economic behavior. It’s actually the primary channel by which the Fed controls the macro aggregates; the concrete Fed actions tend to be in the noise, given the size of the US economy.

  17. Gravatar of Jordan Jordan
    23. February 2013 at 13:52

    Scott:

    I agree with you overall that the great inflation was caused by bad monetary policy, but I’d like to make a couple points.

    1. Is it fair to compare the 1973 oil shock with let’s say the 2000-2008 run up in oil? Since the 73′ shock happened from an unexpected embargo it was indeed a “shock”. I was always taught in microeconomics that the demand curve for oil became more elastic as the time frame increased, so consumers and producers adjusted their consumption habits more easily. In that case would you still consider 00-08 a “shock”?

    2. I think a lot of people who were taught the textbook NK models,including myself, had a tough time conceptualizing how exactly expansionary monetary policy could cause stagflation when it only adjusts the AD curve on an upward sloping AS curve. The way most textbooks show adjustment in the medium run, when money becomes neutral, is that over time prices adjust and the AS curve shifts back to a natural level of output and higher prices, and any more money expansion leads to higher prices, output unchanged. This however doesn’t explain both higher prices and lower output as with stagflation. A professor of mine however explained to me that a better way to show stagflation is on a different AS curve where there is an upward sloping portion for the keynesian short run, which turns vertical(classical) at full employment, but eventually slopes downward at the top, as prices rose to a level where unemployment became “voluntary”, which could explain stagflation and expansionary monetary policy, as any shift to the right in the AD curve would cause higher prices and lower output.

  18. Gravatar of marcus nunes marcus nunes
    23. February 2013 at 16:41

    More ‘pedagigical lies’, from Mishkin:
    http://thefaintofheart.wordpress.com/2013/02/12/just-one-of-those-things/

  19. Gravatar of Is it OK to lie for pedagogical reasons? | Fifth Estate Is it OK to lie for pedagogical reasons? | Fifth Estate
    23. February 2013 at 17:48

    […] See full story on themoneyillusion.com […]

  20. Gravatar of TallDave TallDave
    23. February 2013 at 17:50

    I didn’t realize that’s how that time was taught to most people, the story I had always heard was that (under the prevailing theory of the day that inflation and unemployment were two sides of a coin) the Fed tried to target unemployment and found out that doesn’t work very well, and eventually Volcker came along and cleaned that mess up with the early 80s recession and everyone has worshiped at the altar of inflation targeting ever since.

    Here’s an interesting graph of deficits as % of GDP that illustrates Scott’s point. I was actually surprised, I thought Vietnam was more expensive than that, but apparently we also did a LOT less social spending in those days.

  21. Gravatar of Benjamin Cole Benjamin Cole
    23. February 2013 at 21:09

    Excellent blogging.

    The 1960s-70s inflation was almost purely a monetary phenomenon. Maybe Friedman was right, all inflations….

    BTW, Allan Meltzer goes further, and blames 1960s inflation on social spending, but not Vietnam….

  22. Gravatar of ssumner ssumner
    24. February 2013 at 06:56

    david, That’s all true, but as you note the Austrians weren’t “classical” either. They knew that falling NGDP could cause recessions.

    Geoff, You said;

    “The Fed could engage in OMOs, and the receivers of said money go out and spend that additional money on particular commodities,”

    Why would they do that?

    You said;

    The MM claim that inflation instantly affects the prices of everything is, quite frankly, just absurd.”

    The only thing absurd here is your claim thay MM believe what you claim they believe.

    Jon, Fair enough, but in real terms we ran surpluses during much of the Great Inflation, so it supports my point.

    Ben, You asked;

    “However, if the SRAS curve shifted right during the 1970s, why was their stagflation in the 1970s? Unless I am mistaken, stagflation can only be caused by a leftward-shift of the SRAS curve.”

    The LRAS shifted right at a slightly faster rate. As I noted in my post, a very tiny portion of the inflation was indeed caused by supply shocks, but probably less that 1 percentage point.

    Bill, Excellent comment.

    Jordan, OK, but even the 2007-08 runup in oil prices was huge. The real problem with the oil explanation is that the price of almost everything started rising faster in the mid-1960s, and kept rising fast until the early 1980s. Prices tripled very quickly, even for things like cars (which now rise in price very slowly. Might that have had something to do with wages rising at 10% year in the auto industry? And was that caused by oil? Or easy money?

    Ben, Do you have a Meltzer link? I’d do a post.

  23. Gravatar of Geoff Geoff
    24. February 2013 at 08:00

    Dr. Sumner:

    “Geoff, You said;”

    “The Fed could engage in OMOs, and the receivers of said money go out and spend that additional money on particular commodities,”

    “Why would they do that?”

    Because they’re human. All humans have particular subjective value scales that contain desires for particular goods rather than a desire for a little bit of every single good produced in an economy.

    “You said;”

    “The MM claim that inflation instantly affects the prices of everything is, quite frankly, just absurd.”

    “The only thing absurd here is your claim thay MM believe what you claim they believe.”

    Market monetarists prefer a nominal income target due to their twin beliefs that rational expectations are crucial to policy, and that markets react instantly to changes in their expectations about future policy, without the “long and variable lags” highlighted by Milton Friedman.”

    ???

  24. Gravatar of Geoff Geoff
    24. February 2013 at 08:43

    Don Geddis:

    “@Geoff: Your problem is that you are making assumptions you don’t even realize, which are not shared on this blog (esp. by Scott).”

    You keep saying this, but I have already showed you why it is problematic. It is not a “problem” I have. It’s yours, because you are unable to deal with such disagreements in any way other than cavalier hand waving and evasion. I would very much like to delve into why you are holding yourself up at this stage, and pretending in your mind that just because there are disagreements about assumptions, that there is no recourse other than for people to go their separate ways.

    To help you out, just consider the “assumptions” you yourself are making as you deal with such disagreements and what you are presupposing when you conclude that the disagreeing parties involved should cease and desist interacting. Are there any ethical implications of what you are thinking? Are there any implicit statements you are making when you do that?

    To advise you about what I am thinking, I am not making “assumptions” as if they are stabs in the dark, guesses, beliefs that I am forced to have because I lack the means to establish whether those statements are true or false. That is how religious people “deal” with their disagreements about “assumptions.” It is not how rational, logical, scientific minded people deal with their disagreements. They settle their disagreements via establishing, if applicable, which “assumptions” are false and which “assumptions” are true.

    The “assumptions” I am utilizing are statements that are themselves subject to falsehood or truthhood via self-reflective reasoning.

    I am not really concerned or unsettled by the fact that you can point to disagreements with various “assumptions” that underlie people’s convictions. Did you want me to change my assumptions by faith and simply adopt yours just so that you can understand and agree with the reasoning that I used to get to a particular conclusion? Why do you keep bringing this up?

    “So most of what you say is irrelevant, because you’re using the wrong model. I’ll point out your mistakes, but you should read many of Sumner’s earlier postings on the subject, if you want to understand better.”

    You don’t even understand the model I am using, how can you possibly claim to know it’s wrong? Your entire approach to dealing with these issues is wrong because your epistemology is demonstrably flawed. But in the interests of discussion, I’ll agree to read what you have to say.

    “You write: “The Fed could engage in OMOs, and the receivers of said money go out and spend that additional money on particular commodities””

    “The “receivers” of Fed OMO purchases, have essentially zero change in their behavior, whether or not the Fed engaged in OMOs.”

    Laughably false. If they have no effect, then neither does monetary policy. Monetary policy MUST affect people’s behavior if monetary policy is to affect “the economy.” If it does not affect people’s behavior, then it doesn’t affect “the economy.”

    “Either way”

    Either way? You mean you are granting the possibility that OMOs do affect people’s behavior? That’s a pretty big “either way” after you just got through telling me I am wrong for saying they do have an effect.

    “the receivers used to own some T-bills, and they sold them on the open market, and recieved cash for their bonds. There is no reasonable sense in which those private sellers of bonds, have “more” money because the Fed engaged in the OMO, than they would have if the Fed had not done so. So it’s a complete red herring for you to discuss how the sellers are then going to change their future purchase behavior.”

    If there is no increase in money from OMOs, then the Fed cannot bring about an increase in the money supply or NGDP. Also laughably wrong.

    The bond holders did not sell them in the “open market.” They sold them in a market that includes an institution that can print money. That institution ADDS to the aggregate money supply, and so it is unavoidable, necessary, inevitable that such an institution would be increasing the money that individuals in the market receive as compared to the baseline of no such institution.

    I find it baffling and out of this world how you can actually believe the nonsense that the Fed does not increase the money supply nor does it affect people’s behavior. If you really believed these things, then you must also believe that the Fed cannot bring about the effects that you say derive from Fed activity.

    For some reason there is a gaping hole of reasoning in your worldview. You can identify that the Fed affects the aggregate money supply, aggregate spending, aggregate price levels and other “aggregates”, but you are lost in identifying how to connect those macro concepts to their real world micro foundations of individual people, along with who does what and how in relation to whom.

    If you are unable to figure out how Fed activity affects particular people, then you will not be able to figure out what I am saying such that you can claim that I am right or wrong.

    “Imagine your income doubling.”

    “But that’s not a relevant scenario, because the sellers of the Treasuries don’t find that their income is any (significantly) different, whether or not the Fed does OMOs. So this microeconomic imagining of yours never occurs.”

    Evading a thought experiment is usually a tell tale sign that one’s own worldview is insufficient in dealing with critical analysis.

    You just admitted that incomes are different. You believe you are dealing adequately with it by saying it “isn’t significant.” OK, if you don’t like thought experiments, then just consider your own admission. Imagine your income being HIGHER than it is now. If you are not able to understand the basics of how different incomes have different effects, then you again will not understand what I am saying.

    The point that you need to understand is that when nominal incomes rise, people do not buy a little more of every single thing they used to buy. If you want to deal with smaller numbers, say income rising by 5% per year, then the point is that people will not be spending 5% more on table salt every year, 5% more on mortgage payments, 5% more on gasoline, 5% more on stocks, 5% more on bonds, 5% more on every single possible thing that they can buy.

    What usually happens is that linear increases in incomes are allocated in non-linear ways among goods, services, financial assets, etc. This is why we observe in empirical history something other than every single price rising by exactly 2% or 3% per year. It is precisely why we observe the stock market rising 40% or 50% in a couple years, while electronics prices continue their declining price trend, while certain consumer goods rise in price by 5% while other goods rise by 1%. It’s all different because, as I have pointed out many times, money creation does not bring about a set of human behaviors that lead to more spending on all goods and services equally.

    A higher income will usually be associated with a change to an individual’s valuations of goods and services.

    What you claim is “insignificant” is really just you putting your head in the sand.

    “The Fed boosts whatever particular markets are affected by an increased quantity of money in the possession of the people who receive the new money”

    “Sumner was trying to hint to you, that you’re engaging in a widespread fallacy that it matters “who gets the new money first”. He has addressed this in the past, and has numerous posts debunking the fallacy.”

    It’s not a fallacy! I’ve read those posts you refer to, and they are all untenable.

    It absolutely does matter who gets the money first, because each individual’s value scales are different. New money in the hands of middle class families first, for example, will have different effects than new money in the hands of investment bankers. New money in the hands of the Treasury will have different effects than new money in the hands of stock traders.

    If it didn’t matter who got the new money first, then there would be no reason why any of the banks would be fighting tooth and nail to be on the primary dealer list. There is a very good reason why it pays to be on that list. Again, you can put your head in the sand and believe it is “insignificant”, but there are gains that can be made and they do have effects on the economy as compared to the baseline of, say, everyone getting new money at the same time at the same rate.

    “That is simply not the mechanism by which Fed OMOs raise inflation. It is not a flow of cash, from first point of injection, only later to the rest of the economy.”

    Yes, it is. OMOs add money to the accounts of those who sell bonds to the Fed. Not everyone’s accounts rise in money when the Fed engages in OMOs. Those who receive the new money first have to spend that additional money before others can claim to own more money as compared to the baseline.

    “The Fed sending checks to investment bankers first will have vastly different affects on relative prices as compared to sending checks to low income families first.”

    “Another irrelevant scenario.”

    That was not irrelevant either.

    “The Fed doing OMOs (buying Treasuries on the open market) is a very different operation, from the Fed directly sending “helicopter drops” of new cash to specific individuals.”

    Straw man. I didn’t compare the Fed buying Treasuries to the Fed dropping money via helicopter. I compared the Fed increasing the money balances of bond traders and the Fed increasing the money balances of low income families. How the Fed brings this increase in money about is secondary to my point.

    Once the money is created, and once the sellers send whatever it is they sold to the Fed, new money in the hands of investment bankers will have different effects on the economy as compared to new money in the hands of low income families.

    If you can’t grasp this basic point, then you will not be able to understand my argument.

    “Your new scenario does indeed have distortion effects, in a way that OMOs don’t.”

    I didn’t set up that scenario you claimed I set up.

    “It would require low income families, for example, to increase their spending before their incomes rise in line with the Fed’s inflation, before their incomes rise.”

    “Expectations of future inflation (income & prices), do indeed change people’s immediate economic behavior. It’s actually the primary channel by which the Fed controls the macro aggregates; the concrete Fed actions tend to be in the noise, given the size of the US economy.”

    It cannot change people’s behavior in ways that require them to have more money than they actually do at that point.

  25. Gravatar of Don Geddis Don Geddis
    24. February 2013 at 09:20

    @Geoff: “Monetary policy MUST affect people’s behavior if monetary policy is to affect “the economy.”

    Of course, but it doesn’t affect people’s behavior the way you think it does. It affects people’s behavior, because the overall money supply increases, and thus aggregate demand increases. You seem to think that it differentially affects the behavior of the people who sold the Treasuries directly to the Fed, first, and only everybody else later. Because you seem to think that those first people somehow have higher income. But that’s your critical mistake. Those T-bill sellers don’t have any higher income, just because they happened to sell to the Fed rather than to some other private purchaser. So your idea that the mechanism of monetary policy is by the first sellers having a higher income, is false. That’s not the mechanism by which monetary policy works.

    nominal incomes rise, people do not buy a little more of every single thing they used to buy

    It depends a lot on what happens to nominal prices, at the same time. You seem to be confusing a nominal income rise with a real income rise.

    New money in the hands of middle class families first, for example, will have different effects than new money in the hands of investment bankers.

    And here, you’re confusing “new money” with “more income”. If you sell a T-bill for $100, if the buyer was private, then you have $100 of “old money”. If the buyer was the Fed, then you have $100 of “new money”. But either way, you have the same $100 total. Your behavior doesn’t change, because having “new” money is not at all the same as having “more” money, which is what you seem to be (erroneously) assuming.

    It cannot change people’s behavior in ways that require them to have more money than they actually do at that point.

    And here’s your second big mistake, because people’s economic behavior today can change a lot, depending on their expectations of the future. If you think you’ll get laid off next year, you’ll probably tighten your spending right now. If tomorrow you instead believe that next year you’ll keep your job and even get a promotion, then you’re likely to spend more today, even if your bank account today hasn’t changed at all.

  26. Gravatar of Geoff Geoff
    24. February 2013 at 10:14

    Don Geddis:

    “@Geoff: “Monetary policy MUST affect people’s behavior if monetary policy is to affect “the economy.””

    “Of course, but it doesn’t affect people’s behavior the way you think it does.”

    Except I haven’t even explained in detail how I think it affects people’s behavior.

    I have only said that it does, and that because different people have different value scales, it means the Fed will bring about a particular effect depending on who it sends new money to.

    That’s all I have said. If you agree with this, then we’re done. If you don’t, then what is it that you disagree with, and why?

    “It affects people’s behavior, because the overall money supply increases, and thus aggregate demand increases.”

    No, it’s the reverse. You have cause and effect the wrong way around.

    Aggregate demand increases because the particular increases in money balances of particular individuals increases and because their particular spending increases. An individual person doesn’t spend more just because the aggregate money supply increases. They spend more if their own money balances increase (which is also an increase in the aggregate money supply, but is a consequence, not a cause, for the individual spending more).

    Individuals don’t do more or less because of some aggregate that is allegedly disconnected from individuals. The aggregate accounts are functions of all the various individual accounts. Aggregates change because individuals change, not the other way around.

    You cannot change aggregates without changing individual factors, but you can change individual factors without changing aggregates.

    “You seem to think that it differentially affects the behavior of the people who sold the Treasuries directly to the Fed, first, and only everybody else later.”

    No, I didn’t say that, nor did I seem to say that. I think your problem is that you have a list of false statements in your mind, and your job is to refute them, using me as a means, even if I do not even make those statements.

    My thinking is that it differentially affects the behavior of the people who sold the Treasuries, and it also affects the behavior of those who do not. But I differ from you in that I do not think that the differential effects of the behavior of those who do not sell Treasuries are all instant. I think there is a substantial portion of the population whose behavior is more significantly affected only once the money is actually spent and respent such that the new money finally reaches their bank accounts.

    This is the activity that is a function solely of nominal incomes and prices paid, apart from expectations and estimations. These are the behaviors that are affected by people actually receiving an increased income, which is markedly different from the behaviors that derive from having a probabilistic expectation of a particular higher future income.

    Your behavior of having a probabilistic expectation of a higher income in the future will be different from actually having a higher income now. It would be absurd to deny this.

    Hence, it is wrong to ignore these effects and to focus solely on expectations as if people can predict the future.

    I mean, I hope you can at least realize that just because Bernanke increased the money supply now, it doesn’t mean I know exactly how my income will rise over the next 5 years. It definitely won’t rise instantly, so that alone is sufficient to my point which is that there are effects brought about by “long and variable lags” that MMs ignore.

    “Because you seem to think that those first people somehow have higher income. But that’s your critical mistake.”

    It isn’t a mistake. Your mistake is not realizing what the “higher income” is compared to.

    They do have a higher income as compared to what they otherwise would have had without the existence of the Fed increasing the supply of money. If their incomes were the exact same as compared to the counter-factual of no increase in the money supply, then you would invariably be claiming that the Fed doesn’t bring about a money supply different from what would have otherwise existed if the Fed did not exist at all. Surely you can see how that is a bunch of nonsense.

    If the Fed’s existence brings about an increase in the quantity of money, then it is necessary, it is certain, it is absolutely implied that the Fed MUST be increasing the nominal incomes of people as compared to the counter-factual of the Fed not creating new money at all.

    “Those T-bill sellers don’t have any higher income, just because they happened to sell to the Fed rather than to some other private purchaser.”

    False. If bond sellers only sold into a market WITHOUT a money printer, then prices would of course be lower. With the money printer, the prices they get are higher.

    “So your idea that the mechanism of monetary policy is by the first sellers having a higher income, is false. That’s not the mechanism by which monetary policy works.”

    No, you’re wrong. I am right. You cannot possibly claim that the Fed increases the aggregate supply of money and increases aggregate spending, while at the same time denying that the Fed increases the money balances of individuals as compared to the counter-factual of no Fed and no increase in the money supply. Aggregates are COMPOSED of individuals!

    I refuse to accept your blatantly false worldview that has a giant gaping hole in basic logic.

    “nominal incomes rise, people do not buy a little more of every single thing they used to buy”

    “It depends a lot on what happens to nominal prices, at the same time.”

    No, prices depend on the marginal valuations, not the other way around. You continue to reverse cause and effect. Prices are a function of individual preferences given goods and given money. They aren’t laws of the universe independent of human behavior.

    “You seem to be confusing a nominal income rise with a real income rise.”

    Nope. Just more of you wanting to refute a particular false claim despite what I am actually saying.

    “New money in the hands of middle class families first, for example, will have different effects than new money in the hands of investment bankers.”

    “And here, you’re confusing “new money” with “more income”.”

    You are again evading the point and claiming I am confusing something I am not even confusing.

    Of COURSE new money means more income. If new money did not bring about more income, then the Fed would not be able to increase NGDP, which is a function of all the individual spending activities! If NGDP rises, then it is a necessity that individuals are spending more. Individuals can’t spend more unless they earn more.

    Again, you are seemingly not able to explain that huge hole in your worldview between the macro and the micro.

    “If you sell a T-bill for $100, if the buyer was private, then you have $100 of “old money”. If the buyer was the Fed, then you have $100 of “new money”.”

    Then you just have to include the people who PAID those old money receivers.

    You can’t compare $100 of old money and $100 of new money, if the context is inflation. The comparison would be more like bond traders selling bonds for $90 without the Fed, and $100 with the Fed because the Fed is increasing the money balances of individuals, including bond buyers, who were previously sellers who got more money than they otherwise could have gotten in a market without inflation.

    Can you even explain how aggregate money supply rises and aggregate spending rises on the basis of identifying individual cash balances and individual spending?

    Or is it all magic to you? Aggregates can rise with the Fed, and yet when individuals are considered, they are all equally the same off as compared to if there was no Fed at all? Can you not see how crazy that is?

    “But either way, you have the same $100 total. Your behavior doesn’t change, because having “new” money is not at all the same as having “more” money, which is what you seem to be (erroneously) assuming.”

    No, you don’t have the same $100 total. With the Fed, you would probably have $100 total but without a Fed you would probably only be able to fetch $90, or if you could fetch $100 still, then it means people had to reduce their other expenditures by more than they would as compared to having the Fed there increasing the supply of money.

    Please just consider in your mind how it would be possible for the Fed to increase the supply of money and aggregate spending, without affecting individual bank accounts and individual spending, as compared to a market without the Fed.

    You will find it is impossible to do. But just try, so that you can see how wrong you are.

    “It cannot change people’s behavior in ways that require them to have more money than they actually do at that point.”

    “And here’s your second big mistake, because people’s economic behavior today can change a lot, depending on their expectations of the future.”

    Again you are straw manning me. I didn’t deny that expectations of the future can effect people’s behavior now. I said that these effects are limited. They cannot affect a person’s present behavior in ways that requires them to have more money than they have now. You cannot refute this by straw manning me and claiming I am completely ignoring expectations.

    You are ignoring that which goes beyond expectations. Expectations of the future are not sufficient in explaining present behavior. They are simply not enough.

    For example, if I expected my income to double next year, I cannot double my spending this year. I may be able to increase my spending this year to some extent, but only once my income actually doubles next year, can I do what a doubled income can actually provide. My behavior WILL change next year once my income is actually doubled.

    Now, if you still don’t like thought experiments of “doubled income”. If you prefer to work with smaller numbers, then nothing changes. If I expected my income to rise by 5% next year, then I would still not be able to do today what ONLY an actual 1.05 x Income size can provide me with. I cannot increase my spending NOW just because I expect my income to rise in the future, unless I actually have more money to spend now. If my income has NOT gone up now, then it doesn’t matter if I expect a higher future income. I must wait until I earn that additional income before I can spend that additional income.

    “If you think you’ll get laid off next year, you’ll probably tighten your spending right now.”

    Would I tighten to zero spending today? After all, if my income is going to be zero next year, then according to you, my expectations are all that matters. Oh woops, no it’s not all that matters. I will spend now in part because I have earnings now.

    Expectations do not have a monopoly in explaining current behavior. Current behavior also depends vitally on current conditions (and past conditions). It depends on the future, present and past, not just the future.

    “If tomorrow you instead believe that next year you’ll keep your job and even get a promotion, then you’re likely to spend more today, even if your bank account today hasn’t changed at all.”

    Spend how much more? Suppose I expected a raise of 10%, but I don’t have enough money now to spend 10% more than I do. Here, I cannot spend more now that is solely a function of expectations of higher future income.

    And even if I did expect a 10% raise and even if I could spend 10% more now, it doesn’t mean I will spend exactly 10% more now. I might spend only 5% more now, or less, depending on my knowledge and preferences.

    And if I were living paycheck to paycheck, and I had next to nothing in my bank account, then it would be impossible for me to spend more now, even if I expected a higher future income.

  27. Gravatar of Don Geddis Don Geddis
    24. February 2013 at 12:02

    @Geoff: So many fallacies, I hardly know where to begin.

    different people have different value scales, it means the Fed will bring about a particular effect depending on who it sends new money to. … If you agree with this, then we’re done. If you don’t, then what is it that you disagree with, and why?

    Of course I disagree. From the beginning, I’ve said that the root problem here is that you think it matters who gets the “new” money first. That the behavior of the “new” money recipients changes first (in some way you haven’t specified), before anyone else’s behavior changes.

    You have the wrong mechanism. That’s not how monetary policy affects the economy. There is nothing special about the behavior changes of the people who get the new money first (as compared to the behavior changes of everyone else in the economy, at the same time, as the money supply changes).

    They do have a higher income as compared to what they otherwise would have had without the existence of the Fed increasing the supply of money. If their incomes were the exact same as compared to the counter-factual of no increase in the money supply, then you would invariably be claiming that the Fed doesn’t bring about a money supply different from what would have otherwise existed if the Fed did not exist at all.

    See, here’s where you’re being slippery. The question is whether the immediate sellers of the Treasuries to the Fed have a higher income or not, just because the Fed was the purchaser. You obviously think those individuals do have a higher income. But you’re mistaken. This is not the mechanism of how monetary policy works. It doesn’t work by giving particular individuals higher incomes, and letting them spend those assets to flow to the rest of the economy. That’s just a broken model of how monetary policy functions.

    Now, obviously, people’s incomes throughout the economy do need to increase, you’re right about that. But the critical thing you’re missing is that the increased income does not happen first to the original sellers.

    If bond sellers only sold into a market WITHOUT a money printer, then prices would of course be lower. With the money printer, the prices they get are higher.

    You’re being misled by your microeconomic intuitions. The Fed’s direct effect, by slightly increasing demand for T-bills, on the current market price for T-bills, is trivial. Again, you are mistaken about the mechanism for monetary policy transmission. In fact, the price of T-bills is far more affected by expectations of the economy’s future. So the Fed purchasing T-bills might make the price of T-bills go up, or go down. You make a mistake to assume the same affects as if any other buyer (say, a large hedge fund) were to suddenly purchase a huge supply of Treasuries. The Fed purchases have very, very different effects on the market, than a hedge fund purchase of the same quantity would have.

    Of COURSE new money means more income.

    This is exactly the critical point that you are missing, so perhaps you should think about it more deeply. A particular private citizen sells a Treasury bill to the Fed for $100. Yes, that particular citizen receives “new money”. Yes, the overall money supply increases. Yes, that increased income must appear somewhere in the economy. But where you are wrong, is that the particular citizen who sold the Treasury to the Fed, does not necessarily see his particular income rise. For sure, he has $100 in “new money”. But also for sure, he does not have $100 greater income (than he would have had, if the Fed had not been involved). And yet, the economy as a whole does have $100 greater income.

    But not the specific guy who sold the Treasury to the Fed. That remains your major mistake in understanding.

    With the Fed, you would probably have $100 total but without a Fed you would probably only be able to fetch $90

    So now you seem to have a theory that the Fed’s monetary policy affects the economy, by changing the market price for Treasuries. Another good try, but again wrong. Changes in the price of T-bills is not the mechanism by which the Fed controls nominal aggregates.

    or if you could fetch $100 still, then it means people had to reduce their other expenditures by more than they would as compared to having the Fed there increasing the supply of money.

    Ah, now you’re starting to understand. The behavior and income changes that you’re looking for (obviously required, if NGDP is going to change), can first occur far, far away from the specific transaction of selling a Treasury to the Fed.

    This whole argument is about whether it matters “who gets the new money first”. It doesn’t matter, because “getting new money” is not the mechanism by which monetary policy works.

    I cannot increase my spending NOW just because I expect my income to rise in the future, unless I actually have more money to spend now

    That’s simply not true. There are two obvious mechanisms: for one, you can take out a loan today (e.g. car loan or house mortgage). For another, most people have savings, and they can choose current economic behavior to either save more and increase their savings (if they are concerned about the future), or else spend down their current savings today (if they are optimistic about the future and no longer see the need to have such a large financial cushion).

    The vast majority of the economy does not run paycheck-to-paycheck, always exactly spending the totality of their current income (no more, no less), in any given time period.

    That’s why expectations matter so much. Because people in fact have enormous economic freedom to smooth their consumption patterns over time.

  28. Gravatar of Geoff Geoff
    24. February 2013 at 12:30

    Don Geddis:

    “@Geoff: So many fallacies, I hardly know where to begin.”

    You can start with the umpteen fallacies I have identified in your series of posts, if that suits you.

    “different people have different value scales, it means the Fed will bring about a particular effect depending on who it sends new money to. … If you agree with this, then we’re done. If you don’t, then what is it that you disagree with, and why?”

    “Of course I disagree. From the beginning, I’ve said that the root problem here is that you think it matters who gets the “new” money first. That the behavior of the “new” money recipients changes first (in some way you haven’t specified), before anyone else’s behavior changes.”

    I’ve said since the start that this is not a “root problem”, because the problem is your refusal to accept the basic truth that if the Fed brings about aggregate effects and changes from the baseline of no Fed activity, then it must necessarily bring about effects and changes to individual’s behavior.

    You are simply flat wrong to believe otherwise. Sorry.

    “You have the wrong mechanism. That’s not how monetary policy affects the economy. There is nothing special about the behavior changes of the people who get the new money first (as compared to the behavior changes of everyone else in the economy, at the same time, as the money supply changes).”

    Again, you’re wrong about this. You don’t seem to be able to connect macro statistics to their micro economic realities.

    Of course there are changes to the behaviors of individuals who receive the new money first, for the same reason there are changes to macro aggregate statistics. You can’t have the latter without the former. It would be like claiming you can increase the size of a balloon without affecting any of the individual molecules of which the balloon is composed.

    “They do have a higher income as compared to what they otherwise would have had without the existence of the Fed increasing the supply of money. If their incomes were the exact same as compared to the counter-factual of no increase in the money supply, then you would invariably be claiming that the Fed doesn’t bring about a money supply different from what would have otherwise existed if the Fed did not exist at all.”

    “See, here’s where you’re being slippery. The question is whether the immediate sellers of the Treasuries to the Fed have a higher income or not, just because the Fed was the purchaser.”

    Excuse me, but no, there is no “the question” apart from the specific argument I intended to make.

    Of course individual incomes are higher than they otherwise would be without the Fed. Again, if they were not so, then the Fed would not be able to bring about an increase in the quantity of money or volume of spending beyond what the market would have done on its own without the existence of the Fed.

    Your brain seems to be completely hung up on not being able to connect macro with micro. You can get that the Fed increases various aggregates, but as soon as you try to understand how the Fed affects individual’s behavior and accounts, you get completely lost. It seems like to you the Fed can affect aggregates without affecting any individuals!

    Can you not see how that is absurd? Please address this rather than hand waving at it like you are doing now. Consider what I asked you to consider: How can the Fed bring about changes to aggregates, how can the Fed affect people’s behaviors in the aggregate, if it isn’t bringing about any effects to individual accounts or changing individual’s behaviors?

    “You obviously think those individuals do have a higher income. But you’re mistaken.”

    No, YOU are mistaken. You don’t understand how the monetary system works.

    The Fed doesn’t only affect aggregate statistics. It cannot only affect them without also affecting individual statistics. If it doesn’t bring about any differences to those they deal with, then those people cannot alter their behavior with others either compared to the baseline, and those they deal with then cannot bring about any changes as compared to the baseline, and so on, and all along, there can be no changes to any aggregates either.

    But that’s nonsense. If there are changes to aggregates, then it is necessary that the Fed be affecting the people they deal with as compared to if the Fed didn’t exist at all.

    What you are saying is akin to saying that you can fill up a bathtub with water, using a faucet that does not affect the water height at the point of entry.

    “This is not the mechanism of how monetary policy works.”

    Yes it is.

    “It doesn’t work by giving particular individuals higher incomes, and letting them spend those assets to flow to the rest of the economy. That’s just a broken model of how monetary policy functions.”

    No, it’s the correct model. Your model is broken. If the Fed did not affect the incomes of those they deal with directly, then those people could not affect the incomes of others, and those people could then not affect the incomes of others still, and all along, there would be no change to any aggregates.

    Obviously you’re wrong, and continuing to merely accuse me of being wrong, without actually explaining how I am wrong, as opposed to me who is explaining how you are wrong, is not a very convincing method you have going on.

    “Now, obviously, people’s incomes throughout the economy do need to increase, you’re right about that. But the critical thing you’re missing is that the increased income does not happen first to the original sellers.”

    Astounding. Simply astounding. Here you are being presented with the logical necessity that people’s incomes have to rise, and for some reason you refuse to accept the fact that the people the Fed deals with directly are the only possible group of people whose incomes must rise, IF there is going to be a rise in incomes cause by the Fed throughout the economy.

    If it’s not the receivers of money whose incomes are higher, and the Fed ONLY deals with those receivers, then how in the world can everyone else’s money balances and incomes rise? OBVIOUSLY the rise in money balances and incomes of those who do not deal directly with the Fed DEPEND on those who DO deal directly with the Fed!

    If those who do not deal directly with the Fed depend on those who do for increased money balances and incomes, then it is a logical necessity that the initial receivers’ bank accounts and incomes rise by virtue of the Fed!

    Dude, seriously, this is basic logic here. You cannot possibly deny it without making a serious error somewhere in your reasoning.

    In other words:

    If the Fed only deals with group A, and not ~A, and we know that the money balances and incomes of ~A are rising over time, and the ONLY connection between the Fed and ~A is group A, then it is NECESSARY that the money balances and incomes of group A rise, i.e. are affected by, the existence of the Fed, as compared to the baseline of no Fed.

    “If bond sellers only sold into a market WITHOUT a money printer, then prices would of course be lower. With the money printer, the prices they get are higher.”

    “You’re being misled by your microeconomic intuitions.”

    No I’m not. You are. You are being misled by a psychological refusal to accept the fact that the Fed’s existence benefits a particular group of people who are not you.

    “The Fed’s direct effect, by slightly increasing demand for T-bills, on the current market price for T-bills, is trivial.”

    AHA!!!!!!

    You are either highly confused, or a liar. After claiming I am wrong to believe that the Fed does not affect the money balances or incomes of the initial receivers, you now admit that the sellers of Treasuries experience a “slightly increased demand.”

    I completely wasted my time with you. You’re not being serious.

    “Again, you are mistaken about the mechanism for monetary policy transmission.”

    YOU JUST ADMITTED I AM RIGHT.

    “In fact, the price of T-bills is far more affected by expectations of the economy’s future.”

    So you admit that they are affected!

    I never said they are affected by X%. I said they are affected.

    Can you not see the difference? You are imagining me to be making all these arguments that I am not making.

    “So the Fed purchasing T-bills might make the price of T-bills go up, or go down. You make a mistake to assume the same affects as if any other buyer (say, a large hedge fund) were to suddenly purchase a huge supply of Treasuries.”

    No, YOU are mistaken, not me.

    “The Fed purchases have very, very different effects on the market, than a hedge fund purchase of the same quantity would have.”

    That’s what I have been telling you, on deaf ears. Now you’re telling me this like I don’t know? You’re hopeless.

    “Of COURSE new money means more income.”

    “This is exactly the critical point that you are missing, so perhaps you should think about it more deeply.”

    HAHAHAHAHAHA

    Yeah, we’re done. You need a serious session of economic thinking. What you’re doing is coasting.

  29. Gravatar of Don Geddis Don Geddis
    24. February 2013 at 13:06

    @Geoff: “After claiming I am wrong to believe that the Fed does not affect the money balances or incomes of the initial receivers, you now admit that the sellers of Treasuries experience a “slightly increased demand.”

    You have two problems: (1) “increased demand” does not necessarily mean “increased price” (and/or incomes to the receivers). It depends what happens to the supply curve. Price, obviously, is a function of both supply and demand. This is why I said you are being misled by thinking of something like a hedge fund. A hedge fund buying Treasuries won’t change the supply curve for Treasuries. But the Fed doing so, can.

    (2) Even if the price changes slightly, this effect is trivial. You’re focusing all your attention on this possible trivial effect, but you’re completely missing the huge monetary effect of the change in the money supply. (Along with the expectations channel.) It’s the money supply effect which causes the whole massive economy to change course, not the (possible) trivial change in the market price of T-bills.

    You seem completely unaware of the real monetary policy transmission mechanism, and assume that all of the action must take place through changes in income to the primary dealers. But the truth is, whether the primary dealer income changes are “zero” or “a trivial amount”, it’s completely swamped (in the overall economy) by this other effect, of which you seem to have shown no awareness at all.

  30. Gravatar of Don Geddis Don Geddis
    24. February 2013 at 13:23

    @Geoff: “If it’s not the receivers of money whose incomes are higher, and the Fed ONLY deals with those receivers, then how in the world can everyone else’s money balances and incomes rise?

    Since this seems to be a critical part of your misunderstanding, I’ll try to give you a concrete scenario. Two people, A and B. A is interested in selling a T-bill that he owns, for $100. B is interested in buying a T-bill for $100. Without the Fed, A sells the T-bill to B. $100 is transferred from B’s bank account to A’s bank account. In the overall economy, the money supply has not changed.

    Scenario 2: the Fed enters the picture. This time, A goes to sell his T-bill, but now the Fed buys it for $100. Note that A’s situation is identical to Scenario 1: in both cases, he had a Treasury before the transaction, and afterwards he has $100 in his bank account. (Money is fungible, so he obviously can’t tell the difference between “new” money and “old” money. In either case, he has exactly $100 in his account.) We shouldn’t expect A’s behavior to change, because his situation is identical.

    But in this case, Scenario 2, B’s situation is very different. Because the Fed purchased the T-bill that B otherwise would have purchased. So B instead keeps his $100 in his own bank account, and doesn’t buy a Treasury after all. Instead, perhaps, he decides to buy a car or a meal at a restaurant. The difference is, in Scenario 2, the world economy now has a total of $200 ($100 in A’s bank account, and also $100 in B’s bank account). Overall demand has doubled.

    But the individual change in behavior, happened not on the part of A, who was directly dealing with the Fed. Instead, it was B’s behavior, who otherwise would have dealt with A, if only the Fed had not gotten there first.

    So there you go. A concrete example, of how the Fed’s actions can effect the overall economy (esp. aggregate demand), through a mechanism other than changing the incomes of the primary dealer that the Fed interacts with, which is apparently the only mechanism you can conceive of.

  31. Gravatar of Greg Greg
    24. February 2013 at 15:32

    Don, your reasoning makes no sense. You don’t understand Cantillon Effects either. I’m sure it’s a mere coincidence that all governments in the world are in debt (one of the first touchers of new money is government).

    When Zimbabwe creates all this demand with their money printing, does it create new wealth as well? Or just when the Fed prints new paper money and buys things with it does this magic happen? I use the word magic because that’s what it is when you create things out of thin air like the Fed does. But it creates wealth this way, no? You should really try and grasp what Geoff is saying. You are simply over your head, as is the owner of this blog. He is yet another inflationist bias confirmer who doesn’t want to hear his pet theories destroyed because it will destroy his world view.

  32. Gravatar of Don Geddis Don Geddis
    24. February 2013 at 15:59

    @Greg: Sumner has covered Cantillon effects in monetary policy before. The Austrian concern there is hardly persuasive. It’s not that I don’t understand it. It’s that I reject it.

    You then argue against a strawman, as though any MM proponent is suggesting that printing nominal money is equivalent to real growth. But that’s not at all what Market Monetarist theory says. We all agree that creating wealth, and printing paper money, are different things. And in any case, it’s not relevant to the confusion Geoff is having (which is about whether “new money” has some kind of extra “early” purchasing power).

  33. Gravatar of Philo Philo
    24. February 2013 at 18:13

    It was never made clear what were the supposed “pedagogical reasons.” I see no legitimate pedagogical reason for falsifying economic history.

  34. Gravatar of Is it OK to lie for pedagogical reasons? « Economics Info Is it OK to lie for pedagogical reasons? « Economics Info
    24. February 2013 at 19:00

    […] Source […]

  35. Gravatar of Tyler Joyner Tyler Joyner
    25. February 2013 at 08:41

    Don,

    The second scenario you presented seems like a good example of what Geoff is talking about, because a couple implausible assumptions are made in it.

    First, you assume that despite having a new buyer in the market (not to mention one who can print money), the price of the treasury remains $100. You use this questionable assumption to support the argument that A’s actions do not change based on Fed OMOs.

    Then, you assume that despite the treasury still being priced at $100, person B not only does not buy A’s T-bill, but chooses to buy a totally different good. If the price is the same and person B is the same, why would he not buy someone else’s treasury?

    The only way B’s individual decision-making would actually be influenced by the Fed’s actions, in your scenario, would be if the price of T-bills rose relative to his own income, changing his marginal propensity to buy a T-bill over a steak dinner. If true, that would also suggest that A’s T-bill kept its value and also gained part of the difference between the buying power B used to have, and the buying power he has now.

    I tend to agree that whether A is selling to the Fed or to some third party is irrelevant, but that does not change the fact that sellers of T-bills directly benefit from the Fed’s action in your scenario, and buyers do not. There are winners and losers.

    I think Geoff has raised many good points here, which have not been adequately answered.

  36. Gravatar of ssumner ssumner
    25. February 2013 at 09:19

    Geoff, That quote referred to the prices of financial assets and commodities, which are flexible, not the sticky prices of most goods and services.

  37. Gravatar of Doug M Doug M
    25. February 2013 at 10:19

    “Is it OK to lie for pedagogical reasons?”

    No…well it depends. The entire study of economics is a lie. Economics tries to explain the behaviors of people and groups according to various sets of assumptions or models. It is okay to teach the models. But, it is necessary to identify the weaknesses and failures of those models.

    As a university professor it is more important to teach your students to ask the important questions than it is to give them answers. Especially if those answers are comforting but incorrect. Call out the lies!

  38. Gravatar of Don Geddis Don Geddis
    25. February 2013 at 15:33

    @Tyler Joyner: You make good points, so let me grant that my example was deliberately oversimplified. But I was trying to clearly address Geoff’s (misguided) intuitions.

    Take the year 2008. NGDP growth was somewhere below 0%. Sumner suggests that the Fed should have made it 5% instead, via OMOs (and much clearer expectations). Geoff is concerned about the distorting effects of the OMOs.

    Geoff first worries about the “primary dealers” being “the first” to get “new money”. Even you can see that, for the seller of a Treasury, money is money, and it doesn’t matter whether the money is “new” or “old”. It provides the same purchasing power either way. (Geoff also seems to think that selling a T-bill to the Fed for the market price, is somehow the same as the Fed just handing you additional free income, which is also false.) So Geoff’s intuitive concern is mistaken.

    Then Geoff questions whether the Fed could possible influence anybody’s behavior, without first influencing the behavior of the people they touch directly. So I gave the example of A and B, where the Fed deals directly with A, but it is B’s behavior that changes. So we can see that Geoff’s second intuition is also mistaken, that the influence from the Fed’s actions does not need to flow from the people that the Fed touches directly.

    Now you bring up a more subtle point, as to whether we can get B to change his mind about buying a Treasury, unless the Fed’s actions change the market price. For the moment, let me concede this point. So let’s say, that because the Fed does OMOs in my Scenario 2, the open market price rises from $100 to $101. And this is what causes B to choose not to buy a Treasury after all. He would have bought one at $100, but he’d prefer to buy a car or meal, if the price is $101.

    So now you (and Geoff) worry about the extra $1 that A gets, that he wouldn’t have gotten without the Fed’s intervention. And let’s even grant that we don’t know exactly how A will choose to spend this extra $1, and it’s unlikely that it will be spent evenly on every product in the economy, so now we’re all concerned about Cantillon effects.

    The problem, of course, is that the extra $1 for A, is a trivial part of the effects of the Fed’s OMO. The huge effect comes because the total purchasing power of A and B combined, in terms of their bank account balances at the end, is now $201, whereas without the Fed’s OMO, the sum of A and B’s bank accounts was always $100. Aggregate demand for the rest of the economy has changed hugely, and whether A gets $101 or just $100, is totally in the noise of the effect on the economy of the change in the money supply. Geoff is focusing on this possible trivial change, without realizing the real mechanism for the vast majority of the force of monetary policy.

    But it’s even worse than that! Because even you are thinking of the Fed merely as a very rich purchaser, where the choice to make additional purchases must shift the demand curve for T-bills, and thus raise their price. But you too, are overlooking the money supply effects.

    A Treasury is a promise to pay back current dollars, with more future dollars. Imagine that current T-bills pay 1%, and expected future inflation over the next decade is 0%. This makes owning one a “good deal” for many people, as they will receive greater purchasing power with their future dollars, then they have today.

    Now the Fed starts a program of OMOs. You worry this raises the demand for Treasuries, and thus the price. But at the same time, the increase in the money supply may well raise expectations of future inflation, say from 0% to 3%. Now all of a sudden, owning a 1% T-bill isn’t such a good deal after all. So the overall demand for those particular Treasuries drops, and the price drops. It’s a more complicated scenario than my #2; perhaps A sells and B buys, as always. But now what happens is that additional owners of Treasuries (C, D, E, F…) suddenly decide, given the new higher inflation expectations, that they’d rather have the current cash, than this promise of 1% in ten years. So they flood the Treasury market with additional supply, until the price drops enough below par value that it is finally worth holding a 1% note under conditions of 3% inflation. And of course the Fed happily buys up the additional offered T-bills from C, D, E, F…

    Net result: the total bank balances by A,…,F grow by every dollar the Fed “spends”, and the price of T-bills doesn’t rise at all (drops, in fact), and the money supply grows, and there is no purchasing advantage to getting “new money”, and there are no Cantillon effects (even trivial ones) from the Fed “interfering” in the free market for Treasuries, etc.

    And recall that at the end of the day, we get to aggregate demand: A,…,F eventually decide to spend their bank accounts on something else, say cars. And they show up at the car dealer, and compete with other buyers X,…,Z, who perhaps had nothing at all to do with any Treasuries or Fed actions. But from the perspective of the car market, there are now 26 A,…,Z equal car buyers, and overall demand is higher than it otherwise would have been, but there is absolutely no sense in which any of the A,…,Z car buyers is somehow “privileged” because they were a primary dealer, or got the “new” money “first”, or anything like that. Money is fungible, and it all has the same purchasing power.

    The whole point is: all the power from Fed OMOs comes from the change in the overall money supply, which affects everyone everywhere equally, and it doesn’t matter where the “new money” enters the economy “first”.

  39. Gravatar of woupiestek woupiestek
    2. March 2013 at 15:02

    Lies from teachers can prepare students for lies from employers, governments and the media, but I cannot tell you how to judge the students’ reactions to such lies, let alone how to reflect their performance in their grades.

Leave a Reply