A response to Greg Mankiw – Part 1

On October 2, 2019, I received an E-mail from Gregory Mankiw. It was sent to me, Randy Wray and Martin Watts and asked us some questions about our textbook – Macroeconomics – which had been published by leading textbook publisher Macmillan in March 2019. The book has been selling strongly with a third printing already in the pipeline and a second edition coming, hopefully, later next year. Macmillan also publish Greg Mankiw’s macroeconomics textbook, which has been the dominant teaching book in undergraduate programs. I will take you through the E-mail correspondence that followed because it puts in context what Greg Mankiw decided to do next. Instead of continuing the correspondence on academic terms, which was a reasonable expectation at the time, given the initial approach and our replies, he decided to submit a paper – A Skeptic’s Guide to Modern Monetary Theory (December 12, 2019) – to the American Economic Association meeting in early January, which purports to be a ‘guide’ (meaning in English – a framework to convey an appreciation of something) to Modern Monetary Theory (MMT). After his initial entreaty and our responses in good faith, Greg Mankiw clearly decided that engaging with us on the terms he initially set out was not going to be in his interests and thus took another tack, without any further consultation or reference to his initial contact with us. I wasn’t impressed with that strategy. I was less impressed with the ‘guide’ that emerged. It says very little about MMT. It demonstrates how hard it is for someone deeply locked into a dominant but failing paradigm to think outside the ‘box’ for a while and try to understand that the ideas of a new and emerging paradigm cannot be meaningfully reduced back into the conceptual framework of the failing paradigm that the contender is seeking to usurp. I guess his strategy is understandable – after all – our book is now a direct competitor for his textbook and offers a new approach that has much stronger empirical correspondence. In that context, it is in Greg Mankiw’s self interest to attack our book in any way he can. The problem is that attacks have to have some foundation to resonate. Greg Mankiw’s attack is so lateral that he would have been better to have remained silent. Sure, he is playing to the mainstream groupthink echo chamber. But the echoes will die eventually as more and more people realise the mainstream is in its last death throes. This is Part 1 of a two-part response to Greg Mankiw’s paper. In Part 1, we review the E-mail trail that started all this. In Part 2, I will discuss his response.

The E-mail trail

October 2, 2019

On Wed, Oct 2, 2019 at 4:35 PM Mankiw, N. Gregory … wrote:

Dear Professors Mitchell, Wray, and Watts,

I have been reading your book Macroeconomics. (Congratulations on writing it, by the way. I know how much work such projects are.)

To help me get my head around MMT, I have a question that I hope you can help me with. Here it is:

If you could design an experiment to distinguish MMT from conventional macro theory, what would it be?

What I am looking for is something along the lines of the following: If the government does X, conventional macro theory predicts Y, whereas MMT predicts Z. And Y and Z must be observably different.

The reason I ask is that I am wondering whether MMT offers different predictions or, instead, is a new conceptual framework for understanding conventional predictions. I know that running a macro experiment is likely infeasible in practice, but thinking about it might help clarify the theory for me.

Please let me know if there is an easy answer. And I apologize in advance for imposing on your time.

Greg Mankiw

Which seemed to be a perfectly reasonable way to engage in a conversation. It was seeking input from us which suggested that there would be some conversation.

October 3, 2019

On Thursday, October 3, 2019, Randy Wray replied along these lines (not consulting either Martin nor myself prior to sending it):

Dear Greg: Thanks for opening a dialogue. Yes I think we can come up with a number of such hypothetical experiments.

I’ll just start it off with one.

Government increases spending such that G > T – what is normally called a spending deficit (a flow). The conventional view depends on the model – so let’s first take loanable funds and then ISLM.

a) The additional demand for funds pushes up interest rates and crowds out investment.

b) The additional spending flow raises money demand, pushes up interest rates, and crowds out investment.

I do know that you could come up with special cases where this did not happen but I’m keeping it simple.

The MMT view is that if G > T, bank reserves will have been net credited and hence the pressure on the overnight interest rate will be downward, all else equal.

Again, I am keeping this simple.

So the orthodox approaches predict upward pressure on rates, while the MMT approach predicts downward pressure.

We can then get into central bank reactions to these effects in both the conventional and the MMT approaches, and as well into actual procedures adopted by central banks and treasuries to effectuate spending and taxing operations. (ie there may be operations BEFORE as well as AFTER the sequences I outlined).

So that seemed straightforward and after reading that I assumed Randy was anticipating a conversation would follow in response to this ‘experiment’ which Greg Mankiw had requested.

October 3, 2019

Given the time differences between the US and Australia, my reply came a later (and I didn’t consult Randy or Martin before I sent it). Randy’s response had come overnight in Australian time.

When I had finished work for the next day, I wrote the following (Thursday, 3 October 2019 at 19:07):

Dear Greg (if I may)

Thanks very much for your enquiry and I think the ‘experimental’ approach to differentiating MMT from the mainstream monetary macro is interesting.

I laughed when I read your congratulations – we are definitely fellow travellers in the textbook sense – I never imagined writing such a book would be as difficult as it turned out to be. Much harder than a refereed journal article and/or a regular monograph.

Here are some ideas.

Randy, has provided one good example of the differentiation between the mainstream macroeconomics taught in our universities and the MMT approach.

Here are some more points of departure which can easily be framed as experiments, although in the real world we have already seen these examples play out before our eyes.

The first relates to the existence or otherwise of the money multiplier, which is a central mainstream proposition but is absent in MMT – in fact, we consider the causality runs in the reverse direction.

In mainstream courses, students get taught the money multiplier which says that the central bank determines the supply of loans in the economy by controlling the quantity of base money (reserves).

They learn that there is a (constant) ratio of broad money to base money, and when the central bank adds reserves, they are multiplied up to create a greater change in bank loans and deposits.

The theory has two elements:

1. Reserves constrain lending – a supply-determined approach.

2. Central banks control reserves.

MMT says that:

1. Reserves do not constrain lending – loans create deposits (not the other way around) – a demand-determined approach

2. The central bank does not control the amount of reserves.

3. Banks seek credit worthy borrowers – then create loans which create deposits.

4. The central bank will always supply reserves to banks seeking them – so unlike the Money Multiplier story – the base follows the broad (which is driven endogenously by demand for loans from credit-worthy borrowers).

5. The banks cannot clear an excess of reserves in the interbank market (this feeds into Randy’s point about the relationship between deficits and interest rates).

The causality is the reverse of the mainstream approach.

Thus, at the empirical level, mainstream macroeconomics struggles to explain why the broad monetary aggregates have not grown commensurately (as if multiplied) with the dramatic expansion of the asset side of central bank balance sheets.

MMT has no problem explaining that phenomenon.

Another example that follows relates to inflationary consequences of the expansion of the central bank balance sheets.

Mainstream theory predicted inflationary consequences arising from Quantitative Easing (QE) because they considered the increase in base money would feed in, via the money multiplier, to broad money, which, via the Quantity Theory, would drive up prices as bank lending accelerated.

MMT predicted that bank lending would not accelerate as a result of QE because the sluggish lending was not due to a deficiency of reserves (banks do not loan out reserves) but, rather, a shortfall of creditworthy borrowers due to the uncertain conditions following the GFC.

Further, MMT constructs QE as an asset swap only which may reduce interest rates in the relevant maturity range but has no direct implications for the growth of broad money.

Mainstream theory is problematic when trying to understand why the rapid expansion of central bank balance sheets did not cause inflation.

And relatedly, is the question of central banks and government bond yields.

Two points of differentiation.

1. Central bank purchase of government debt

Empirically, central banks have purchased significant proportions of government debt in recent years. The Bank of Japan now holds 43 per cent (approx) of all outstanding Japanese government bonds.

The ECB has a rising share as does the US Federal Reserve Bank.

Mainstream theory considers this to be ‘money printing’ and predicts it would be inflationary.

MMT indicates the inflation risk is in the spending choices of government (and firms, households etc) and not the monetary operation that might be associated with the spending.

In the standard mainstream government budget constraint rendering, governments have two choices to ‘fund’ their deficit spending: (a) bonds, (b) ‘money printing’. (a) drives up interest rates (and crowds out), while (b) is inflationary because it drives down interest rates and expands reserves.

MMT predicts that there is no difference in the inflation risk of the deficit spending arising from these ‘funding’ choices – noting that MMT does not consider the government to be financially constrained anyway.

So issuing debt does not reduce the inflation risk of the spending as the funds to purchase the debt were not being spent anyway.

2. Bond yields

Mainstream theory predicts that if the government increases its stock of debt as a result of running continuous deficits, bond markets will demand higher yields to compensation for the increased risk and higher expected inflation.

Take Japan as an example.

It has had continuously large fiscal deficits and QE programs which have resulted in a massive buildup of excess reserves in the banking system, yet interest rates and bond yields have been low (and sometimes negative) for several decades.

The empirical reality is the opposite to that predicted by mainstream macroeconomics.

MMT demonstrates that if the central bank leaves excess reserves in the cash system (as in the case of the Bank of Japan) then the interbank competition will keep the short-term rates low.

Further, MMT shows that the central bank can control all bond yields in any maturity range if it so chooses. This is exactly what the Bank of Japan has been doing for decades now.

Many financial market investors, following mainstream macro and monetary economics, moved their funds out of bonds and into real assets during the GFC because they expected yields to rise (and capital losses to be made) and inflation to accelerate.

They lost out on both accounts.

MMT predicted they would lose.

Think about the famous ‘widow maker trade’ on the 10-year JGB, which was based on mainstream theory predicting that yields would have to rise as the debt ratio rose.

The short-selling trades all lost out badly.

MMT predicted that the yields would not rise if the central bank chose to control them. The bond markets are supplicant in MMT rather than the drivers of yields.

And, how might mainstream economics account for the fact that:

1. Japan has the highest gross public debt to GDP ratio.

2. Yet, its interest-servicing ‘burden’ is about to be negative overall – that is, investors will be paying the government to hold their long-term bonds. This is because as positive-yielding assets mature, they are increasingly being replaced by negative-yielding issues, such that soon the overall debt servicing burden will be negative.

MMT explains this easily using the logic outlined above.

I hope these ideas help in your thinking about our work.

I could give more examples but time is short today.

I hope one day we might run into each other and have a chance to talk about these things in more detail.

All the best

October 4, 2019

On Friday, 4 October 2019 at 00:41, we received Greg Mankiw’s response:

Thank you for the quick replies. It will take me some days to think through all this, but I wanted to express my appreciate for your taking the time to address my question.


Which all seemed reasonable and I formed the view that he was considering our ‘experiments’ and would respond with his view about them, which would elicit further discussion.

That might seem optimistic to someone without direct professional experience in the Academy, but that is the normal way things pan out in my labour market.

Discussions occur between academics that never see the light of day and when they do are usually acknowledged in the credits of any paper or monograph produced.

Recognising that prior discussion and to-and-fro has occurred doesn’t imply that everyone agrees. But when an academic, as is standard practice, thanks others for their input, they are just recognising that there has been a discussion, or a person has read something or stimulated something that ended up in the final paper.

No more correspondence was entered into (as they say).

Then last week, he published his paper. I only was alerted to it on Twitter by an MMT follower.

Greg Mankiw had decided to have no further interaction with us for his own reasons.


In Part 2, we will discuss Greg Mankiw’s ‘guide’ or non-guide, as the case may be.

That is enough for today!

(c) Copyright 2019 William Mitchell. All Rights Reserved.

This Post Has 26 Comments

  1. Thanks Bill. I went over his paper just very fast and I found several mistakes in It instantly. I am not going to write about them here since you are going to write about It in your blog. He hasn’t even grasped your point about the money multiplier.

    Years ago when he wrote an article about US future, where he described the IMF to be in China, I sent him an email where I pointed to US monetary sovereignty. He sent a reply but It was just meaningless politeness.

  2. I used to think Bill Mitchell was wrong about most of the ‘mainstream’ economist’s being unwilling to listen or learn about MMT. Kind of thought Bill was a little paranoid almost. I admit I was wrong about that. There is definitely something wrong with most in that profession and Mankiw provides more evidence of that with his pathetic ‘Guide’ to MMT.

  3. Prof. Mankiw writes ” “I wanted to identify the key differences between this new approach
    to macroeconomics and the approach in mainstream textbooks”.

    For me, the differences arise out of two key ideas. First, money is a promise (the stuff we call money is how we keep track of certain financial promises). Second, a monetary economy where goods and services are exchanged for promises is materially different than a barter economy where goods and services are exchanged for other goods and services.

    It’s worth noting that humans have been exchanging goods and services for promises throughout recorded history and beyond.

  4. I’m starting reading this very late – should go to bed. It’s too delicious and needs to be savoured – off to beddy byes. Can’t wait for morning. I bought the Mankiew tome many years ago.

  5. Sad state of affairs Bill.

    What I can’t believe is if Mankiw read the MMT textbook how on earth could be not see the clear differences himself ?

    Thousands of us around the world with no university education can highlight these differences. It does not take a rocket scientist to see them.

    Any economics student anywhere in the world could highlight the difference between mainstream and MMT after reading the MMT textbook and sitting a mainstream economics course.

    Yet, here we are again with a giant of the mainstream saying they get MMT and then misrepresent decades of work from MMT economists. They have all done it one way or another and even then after we have highlighted their mistakes in economic thinking. No apology and their narrative changes to They knew it all along. Case in point Krugman on how banks work in reality, he simply had no idea. This was a noble prize winner.

    You nailed this a long time ago Bill.

    ” You’ve done very well at secondary school and then go to university and shine through and achieve academic progress. The selction process as you go through university is that by the time your at the end of your undergraduate years. The bright sparks go through their post graduate studies and a PHD and they get an academic job the creme da le creme.

    So you’ve invested a lot of your early years to get this PHD to get into the academy. With economics you end up with a very defined set of work tools and foregone a lot of income to get to this stage.

    Not many academics become top line researchers most of them become text book pushers. So by the age of 45 you’ve risen to a certain rank in the academy and normally that’s the end of your progression. Good researchers can move up further and get a chair and get to be called a professor. Where a lot of people have reached their ceiling and pump out stuff from a textbook that the publishers brings around every year. Fill up your time as an administrator or a teacher.

    The end of life syndrome approaches and all you are waiting for then is to retire on your pension and live the life and hope your health holds. So why on earth at that point would these people abandon their life’s work and admit that large parts of it are wrong. What’s the motivation for that. Particulary when you are in a group that have incredibly rigid rules with respect to promotion, with respect to asigning status, with respect to getting any publications or research money. In a very disciplined community in economics.

    You’ve learned to play the game and jump through hoops along the way. You’ve learned not to rock the boat. You’ve reached your career progression and waiting for your retirement.

    To come out then and say alot of the textbook pushing has been wrong is to defy your scholastic community. Stop the brightest when they are young from getting promotion, publications and research money. The group membership becomes your priority and when new imperical evidence presents itself to show the textbook pushing has been nonsensical. They’ll forego the oppertunity of a revision of their ideas as maintaining membership and status within the group is more important.

    Groups work out all sort of ways to behave like that. When anomolies come in from the real world they revise history. They rewrite history to reflect the group. Look at all of the revisions carried out after the great depression. Some of stories now being told about the great depression do not reflect reality. they also just deny things and make stuff up. Deny that the whole financial system was saved in 2008 by global government deficits. That’s how they overcome their own personal doubts and maintain membership of the group.

    They are never going to admit their life’s work, after being accepted by the GROUPTHINK royalty is built on some myths Bill. Even if the myths were only in specific areas of their work. It just ain’t going to happen which is why Twitter is a complete waste of time.

    You might as well ask the big names of the mainstream to dunk for Apple’s in boiling water. They would rather do that than betray their group. The ruling class use fake progressives to silence true progressives. It’s like the owners of a prison who order “trustee” inmates to torture regular inmates. If the “trustees” don’t torture, they are tortured themselves after being thrown to the regular inmates. The liberal establishment’s function is to protect the rich and certain mainstream economic professors are playing their part of “trustees “

  6. “The MMT view is that if G > T, bank reserves will have been net credited and hence the pressure on the overnight interest rate will be downward, all else equal.

    So the orthodox approaches predict upward pressure on rates, while the MMT approach predicts downward pressure.”

    This is an excellent example of operations, the kind of detail that mainstream economists like to subordinate to reality.

    The Canadian authorities seem to have a good grasp and put it simply this way (https://bdp.parl.ca/sites/PublicWebsite/default/en_CA/ResearchPublications/201024E#a8)

    “3.2 Cash Imbalance Created by a Fiscal Deficit

    In contrast to a fiscal surplus, a fiscal deficit results in an influx of cash into the private economy19 since the government injects more money through its spending than it collects in taxes. This added liquidity – in the absence of any intervention by the Bank of Canada or the Government of Canada – would create an imbalance in the form of surplus liquidity in the private economy that would drive the overnight rate lower. It is the Bank’s role to neutralize, on a daily basis, this injection of liquidity with a corresponding withdrawal of cash from the private economy.”

    And how did this work in practice on the other side of the deficit/surplus ledger? ie. when Canada had surpluses?

    “Appendix – Fiscal Surplus in Canada: Where did the cash go?

    For 11 consecutive years, beginning with the 1997-1998 fiscal year, the Government of Canada generated a fiscal surplus. One question naturally arises: how did the Government of Canada or the Bank of Canada neutralize the withdrawal of liquidity from the economy that resulted from the fiscal surplus?1 An analysis of the Government of Canada’s balance sheet between 1997-1998 and 2003-2004 showed that a significant portion of the budgetary surplus ended up as government cash reserves rather than being used to pay down debt. Assets in the form of “cash and cash equivalents” on the Government of Canada’s balance sheet actually increased from $10.2 billion in fiscal year 1996-1997 to $22.7 billion in fiscal year 2006-2007. If all this extra cash had been left sitting in the Receiver General account without any intervention from the Bank of Canada in the form of a cash injection into the interbank market, a liquidity shortage would have been created in the private economy that would have pushed up the overnight rate. Of course, the Bank of Canada did not stand still.

    The approach employed by the Bank to neutralize the withdrawal of liquidity generated by the fiscal surplus – more precisely, the portion of the fiscal surplus not used to repurchase government debt – was to re-inject the increased cash reserves from the Receiver General account into the private economy through an auction with financial institutions in the LVTS. This approach is explained by the Bank in its publication The Implementation of Monetary Policy in Canada:2″

    Once again, operations. And not something that is debatable within the confines of irrelevant theory.

  7. I just read Mankiw’s bit.

    One of the basic issues in my view is the commitment to the perspective that divides the world into “natural” and “optimal” conditions.

    The “natural”, a metaphor signifying a self-emergence from its own essence, necessarily creates a “left-over” portion of unused resources.

    The “optimal” is a subjective preference emerging from the wishes of fallible human beings, which is to fully utilise all resources and make the world better.

    So it follows that human efforts to push for the optimal adds a level of subjective “interference” into the “natural” state of things. MMT people may have good hearts but they are engineering into the fragile balance “given” to us by market ecology.

    I think it’s fair to say that the market is not nature, it is instituted (as institutional economics has always pointed out) and therefore this division is bunkum.

    There may be different problems at 90% and 99% resource utilisation, but no ontological difference.

  8. You carefully explain why deficits place downward pressure on the target interest rate and why the excess reserves do not cause inflation and yet Mankiw proceeds to assert the direct opposite without any evidence except resort to an unspecified “wealth effect”. Ultra lame on his part. If he was truly interested in learning about MMT, he could have continued to correspond directly with you or better still read and then cited your copious research like any responsible scholar would. No big surprise about his integrity as he couldn’t even pass his own students’ smell test.

    BTW, great find by Dr NRL on the BoC paper. About as clear as it gets. Proves they all know what’s going on but would rather scare an unsuspecting public with their old time religion.

  9. Well, at least their criticisms have gone from absolutely insanely wrong to just mistaken. I really can[‘t fault someone for echoing Kalecki’s misunderstanding of how central banks treat reserves.:
    “Professor Mints suggests financing the budget deficit by issue of currency in order to avoid the increase in the interest on public debt. Presumably, the government would ob¬tain the currency from the central bank by way of interestless loans. A large part of the currency issued would find its way into the private banks, but the banks would be under the obligation to sterilize the additional cash. The procedure would have two repercussions where there is a substantial long-run “deflation¬ary gap” and thus a substantial permanent budget deficit:
    (a) The “cash ratio” of banks would rise continuously. This would make the banks in-crease the charges on demand doposits and re-duce the interest rate on time deposits. In fact, the increase in the cost of servicing of the pub¬lic debt would not be eliminated but would be, at least partly, levied by the banks on the de-positors.
    (b) The rates of interest would fall continu-ously as a result of the continuous increase in deposits in relation to other liquid assets in the hands of the public. This would cause a decline in the yield of the assets of the banks as the re-spective securities or loans mature. This in turn would make the banks increase the charges on demand deposits and reduce the interest on time deposits still further.”

  10. Userfriendly,
    Kalecki did not have the MMT book on hand in 1946. Mankiw did and was apparently trying to or pretending to use it for his “guide” to MMT. Kalecki was not trying or pretending to describe MMT. There is a big difference.

  11. Not in anyway surprised by Mankiw’s response. You basically disabused him of anything that actually matters, to the point he could no longer sit properly…

  12. I suppose it’s to be expected if they (the academy) misrepresented and bastardised Keynes, they’re going to attempt to misrepresent and bastardise MMT. The important difference this time is that MMT can be understood by us laypeople, so we can see their bullshit for what it is. They’re obviously starting to feel very threatened by MMT, which is excellent!

  13. I replied (as follows) to Greg’s article myself, because I am always annoyed by the sort of line with which he began:

    Dear Sir,
    Your beginning statement (in “A Sceptic’s Guide to MMT”) namely:

    “To be sure, a currency-issuing government can always print more money when a bill comes due. That ability might seem to release the government from any financial constraints”.
    ….”might seem”, but does not; or rather, more accurately in MMT parlance, does not release the government from real resource constraints including labour.
    You then proceed to discuss interest on debt and reserves; but in fact there is no need for government to sell interest bearing bonds to the private sector at all.
    [As to the creation of money and its ‘value’: imagine an (hypothetical) economy in which government has access to sufficient data to entirely plan production and distribution of vital goods and services, *without recourse to money at all*…..to understand that, unlike resources, knowhow, and labour, money has no ‘intrinsic’ value;
    …..in any case, central banks ought to be able to directly create funds for specific public purposes, to supplement money creation in the private sector (when private banks write loans – create deposits – to credit-worthy customers), given the variable conditions in the private sector.
    Provided of course the economy has the productive capacity and spare resources including labour, which is where the MMT Job Guarantee comes in. (end)

    A no doubt simple-minded response by a non-economist to a professor.

    But Keynes himself said (and I will take him at his word) in the preface to his magnum opus: “General Theory of Employment, Interest, and Money (1936):

    “The composition of this book has been for the author a long struggle of escape, and so must the reading of it be for most readers if the author’s assault upon them is to be successful,-a struggle of escape from habitual modes of thought and expression. The ideas which are here expressed so laboriously are extremely simple and should be obvious. The difficulty lies, not in the new ideas, but in escaping from the old ones, which ramify, for those brought up as most of us have been, into every corner of our minds”.

    [With acknowledgements to professor John Harvey, for his excellent article refuting Krugman, Summers and Rogoff, entitled “MMT: sense or nonsense”].

  14. I would personally think about Prof. Mankiw’s article in a more positive way, as an opportunity and invitation to a more serious debate.

    Kalecki’s views are not inconsistent in principle with the statement that running budget deficits financed by issuing currency would lead to the fall of interest rates but they were expressed in a rather confusing way. If we look at the whole paper, the missing bit in Kalecki’s thinking was consumption out of stock of wealth. But this was written in 1946! Kalecki would rather tax capital to keep the budget balanced in the long run. I think he had a valid point as stimulating the economy without redistribution of income could reduce unemployment but would not eliminate hoarding real wealth in the hands of 1% or rather 0.1%. Another issue is that in 1946 central banks would not pay interests on reserves.

    Regarding Prof. Mankiw’s paper I think that the only statement I would agree with is that “the ever-expanding monetary base will have further ramifications. Aggregate demand will increase due to a wealth effect, eventually spurring inflation.” Bingo! This is exactly what the money was “printed” for. The first reason was to stimulate via Keynesian multiplier. The second was to increase the (net) wealth of the private sector so that people’s needs for saving would be satisfied. The government does not print money just for the sake of printing. The inflation does not need to be “spurred” if “printing” or whatever term is more politically correct, ceases or moderates early enough.
    The outcome is a result of assuming that consumption of households depends on their expected stock of wealth what includes the stock of money (including the printouts from the government). Exactly as demonstrated by various models in Godley and Lavoie Monetary Economics, where a Modigliani-like consumption function is assumed. But this result which I would call a “hypermultiplier” has nothing to do with Quantity Theory of Money.

    Unfortunately the rest of the reasoning presented in Prof. Mankiw’s paper is a bit disappointing. Why would a fall in interest rates stimulate the economy to the point of creating accelerating inflation? Has it stimulated in 2008? Oh it was a liquidity trap or whatever. Maybe we are trapped in liquidity all the time? Who knows? And interest rates cannot be seriously negative. They will fall to zero and that’s it. This is about “LM”. Now about “IS”: I have found a couple of econometric papers showing very low impact of interest rates on aggregate demand, as long as lending rate stays below so-called “hurdle rate”.

    I would leave the Laffer curve for seignorage where it belongs to – on the napkin.

    How could a government default on printed money? I have heard about defaulting on bonds but seriously, how can a government default on currency? Refuse to exchange a 10 dollars note for 2*5 dollar notes?

    “In U.S. decadal data since 1870, the correlation between inflation and money growth is 0.79”
    And correlation does not mean causation. The same result comes out of Godley and Lavoie “Growth” model. Crank up inflation (due to a distributional conflict) and the stock of money will grow in the long run. Hardly surprising.

    Seriously speaking the idea of “demand for money” is a common misinterpretation of what Lenin, sorry, Keynes, said. He wrote: “The primary effect of a change in the quantity of money on the quantity of effective demand is through its influence on the rate of interest” . What if the rate of interest is already zero?

    “Mainstream theories of inflation emphasize not class struggle but excessive growth in
    aggregate demand, often due to monetary policy.” – no econometric arguments have been presented to support this statement.

    “In the Keynesian regime, unemployment arises because labor demand is insufficient to ensure full employment at prevailing wages; the demand for labor is low because firms cannot sell all they want at prevailing prices; and the demand for firms’ output is inadequate because many customers are unemployed. Recessions result from a vicious circle of insufficient demand.”
    What if the demand for firms’ output is inadequate because of income distribution, that the rentiers and capitalist are hoarding wealth and their income is so high that they don’t know what to spend money on? This determines the point of effective demand. It can be nudged higher by running higher budget deficits. Secular stagnations result from a vicious circle of rising inequality further eroding bargaining power of organised labour.

    “When firms have market power, they charge prices above marginal cost, so they always
    want to sell more at prevailing prices. In a sense, if most firms have some degree of market
    power, then goods markets are typically in a state of excess supply” – fine, this sounds almost like Marx and his prices of production including costs-prices, resulting from the process of transformation of surplus value into profits.

    “This theory of the goods market is often married to a theory of the labor market with above-equilibrium wages, such as the efficiency-wage model.” – this statement is actually flawed as reducing wages would also reduce aggregate demand due to higher marginal spending propensity of workers. We can go after the statement that in equilibrium wages would be equal to marginal product of labour by looking at the tautological meaning of the logarithmic partial derivatives of Cobb-Douglas production function. Just follow Anwar Shaikh’s “Laws of Production and Laws of Algebra: The Humbug Production Function” . Revenue = wages + profits. Wage and profit rates are exogenous parameters. They are not determined by real production function even if one has a similar form (most often it doesn’t).

    I would leave the “natural level” of whatever for Bill to feast on.

    “If price setters are thwarting that goal by raising prices, policymakers can fix that problem by using price guidelines or price controls.” – this would be profoundly stupid and I am not sure where these ideas come from, I don’t think that Venezuela and Iran are good examples of countries implementing MMT recommendations. Price controls were tried and tested in “real socialism” and the policy led to market disequilibrium, shortages of goods, rationing, “inflationary overhang”, rioting and capital flight. There may be exceptional circumstances such as a war, natural disaster etc. If there is a company abusing monopolistic position then there are legal venues to go after that company. These were hardly used in recent times. Or if it is a natural monopoly, just nationalise it.

    “I agree that the government can always print money to pay its bills. But that fact does not free the government from its intertemporal budget constraint.” – this sounds like Loanable Funds Theory v.2.0 (Woodford). Worth debunking another day.

    (disclaimer. these are my views and I may be totally wrong or confused)

  15. Actually I have just realised something.

    Let’s say we are using a Cobb-Douglas aggregate production function expressed in nominal terms with constant returns to scale.

    We think that in equilibrium with perfect competition on the labour market wages should be equal to marginal product of labour but in the model we assume a certain level of monopsony which would drive the wages above the level determined in the first case. So we proceed with looking at prices, wages, aggregate demand etc and in the end the labour market will not clear.

    We may have made a serious error in reasoning. The coefficient in Cobb-Douglas production function expresses the share of profits in output as demonstrated by Anwar Shaikh. If the wages have risen for a whatever reason as a fraction of nominal output, the coefficient in Cobb-Douglas aggregate production must have changed and we must stop using the original one. It is algebraically impossible in a static model with a Cobb-Douglas production function with constant returns to scale to have wages which are not equal to marginal product of labour assuming the usual conditions.

    I may be wrong but if I am right, we have got an interesting result which has implications to the validity of DSGE models …

  16. This is just one more confirmation that MMT has moved center stage and cannot be stuffed back into the box. The current orthodoxy is misleading and it would be wise for the existing academic bureaucracy to simply acknowledge some mistakes have been made and should now be expunged. And then move on. This is practical and otherwise a good thing. That admitted, the real culprit and the real barrier is that the current orthodoxy ended up being simply a more clever way to funnel wealth to the ruling class of elite wealth and power. So MMT requires real word, systemic changes, beyond the purely academic. That’s a tough nut to crack. As economist Michael Hudson observed, when ancient civilizations forgave debts to re-balance their societies, “there may have been some fighting involved.

  17. Derek Henry wrote:-
    “This was a (Nobel) prize winner”.

    Just highlights once again that the award of the “Sveriges Riksbank’s prize in the economic sciences” (an oxymoron) “in honour of Alfred Nobel” is just a threadbare confidence-trick. Who *did* they think they were kidding, I wonder – other than an inevitable, though small, minority of gullible people (human gullibility knowing no limits;- flying saucers, “miracles”, etc etc).

    What an insult to Alfred Nobel BTW, who was no mean scientist (a chemist) in his own right, to have the gall to “correct” posthumously – ie after he was no longer around to refute them – his deplorable lapse in not having acknowledged economics to be a science.

    How that must have been rankling for it to have spurred them into engaging in such a meretricious gambit as grafting their prize spuriously onto the list of prizes given for achievement in the real sciences!

    Mankiw seems to have found his natural home, along with Summers, Rogoff, Krugmann and others of their ilk.

  18. As a retired litigator, I was struck by how Mankiw’s purported outreach for information to help resolve a dispute was, in reality, only what is called a “discovery” procedure, the ostensible purpose of which is to uncover the factual background of a case, the actual purpose of which is to obtain ammunition from the other side which can be used to weaken the opposing case or bolster one’s own. It does not take many years of law practice to reach the conclusion that the adversarial legal system is not a truth-seeking process but rather an aggressive, competitive, no-holds-barred arena in which one side, for reasons of self-interest, seeks to tell a story, however related to or remote from underlying facts, which will prevail over the story told by the other side–and prevail at any cost, by fair means or foul. Genuine facts come into play only when they can be selectively marshalled to enhance one’s position; otherwise, they are to be distorted, glossed over, deflected, or explained away. Accordingly, what ultimately matters in litigation is not what the underlying facts of a dispute may be, but rather which story being told about it comes across as most believable, often on an emotional or ideological level, to the trier of fact, whether judge or jury. As in law, so in academia, and now under neoliberalism, so in society across the board.

  19. very strange behaviour ,

    why start a conversation, shut it down , and then proceed to try and run over the people you reached out to. one would have thought a reasonable attempt would have been made to continue the dialogue back and forth before any conclusions were reached.

    not quite sociopathic behaviour but troubling.

    believe it or not i think this disappointing exchange is good news

    i think there is hope with mr mankiw, so bill et al , dont give up on him.

    he has read your book, and is now mired in self doubt even though his ego wont admit it.

    he may be having the academic equivalant of an existential crisis

    he is presently at denial. 4 more stages to go.

  20. I`ve only just finished reading ….”A skeptic`s Guide to Modern Monetary Theory” (Mankiw).I wanted to read it before looking at the replies from Bill and Randy. I was tempted to read this post first but chose not to.I did not want to be guided any more than my limited MMT knowledge would take me;
    I say this without any mental reservation whatsoever.
    Read the last three paragraphs, over and over and over again.(forget everything else) The only possible conclusion is that Mankiw does not at all! get it; hints at legitimacy through the final paragraph which sadly, serves only to confirm paragraph four of the paper.

    “At the outset, I should admit that I found the task of figuring out MMT to be vexing. As I
    studied it, I was often puzzled about what precisely was being asserted. I hasten to add that the
    problems I had could have been of my own making. Perhaps after forty years in the profession, I
    am too steeped in mainstream macroeconomics to fully appreciate MMT. I raise this possibility
    because MMT proponents may say that I missed the nuances of their approach. But what follows
    is my honest reaction to MMT after a sincere effort to understand it”

    That is my honest and sincere reaction………I will now read the replies.

  21. more on Prof. Mankiw’s paper

    “in our current monetary system with interest paid on reserves, any money the government prints to pay a bill will likely end up in the banking system as reserves, and the government (via the Fed) will need to pay interest on those reserves”

    Why would it need to maintain a positive real interest rate on reserves? (I am happy if nominal interbank interest rate is equal to CPI inflation)

    To control inflation? But this does not work that way.

    The question about the sensitivity of inflation to interbank interest rate is a question which needs to be tackled from econometric point of view but not by using biased DSGE models which are then “evaluated” and shown to be better than the null hypothesis of random walk. (“we cannot reject the hypothesis that the model’s response better approximates the data than the noise”) And then we have a replicability crisis mentioned in the latest Prof. Quiggin’s paper. DSGE models generate dampened second order ODE responses similar to a dampened pendulum or an RLC electronic circuit. They are not capable of generating “true” long-run hysteresis, unlike SFC models where path dependency is an “inbuilt feature”. If we have a limited time window we can approximate any kind of dynamic impulse response by tying together several first or second order filters. (I can give references if required). And in real econometric data we also have a lot of noise.

    As demonstrated in the paper Bill already commented on, Dynamic Principle Component Analysis which is kind of unbiased except for the choice of variables, clearly shows that the interest rate is not the dominant independent variable affecting other parameters in the business cycle. Aggregate demand is. Fiscal policy directly affects aggregate demand. Monetary policy may kind of have certain impact. Obviously if someone (I am thinking about Prof. Balcerowicz in the late 1990s) whacks the economy with 8%-10% real interest rate, this works as intended, dampening business activity and leading to high unemployment (which peaked at 20%). In the end CPI inflation rate has been reduced.

    The point I am trying to make is that Prof Mankiw implies there is a risk of explosive public debt trajectory and accelerating inflation if the government keeps running budget deficits financed by “printing money” I would strongly disagree with this thesis. If we assume that aggregate consumption positively depends on the stock of wealth (even in a non-linear way), as long as some fraction of the public debt (including the currency, treated as government’s liability) is owned by non-foreign households, the trajectory of public debt / GDP will stabilise. There is absolutely no need to stimulate the economy if the private sector has increased its consumption and firms respond with an increase in capacity-generating investment.

    I think the key argument made by Prof. Mankiw is related to misunderstood in my view “liquidity preference”.
    “if sufficient interest is not paid on reserves, the expansion in the monetary base will increase bank lending and the money supply”

    How precisely the increasing quantity of reserves can increase bank lending? QE was a experiment showing that bank lending does not depend on the quantity of reserves. It does depend on the demand which may be influenced (to some limited extent) by the interest rates. The signal is a price not a quantity. Bank lending is capital constrained but in the medium run banks have no problem raising more capital in the Baron Munchausen’s way.

    Essentially if middle class and rich households don’t want to spend excess disposable income, there is no way they can get rid of the balances ever increasing when the flow is adding to the stock. We would need to draw linked balance sheets of the households, firms, banks, central bank and the government. The only way the households can get rid of the balances is either by repaying the debt or by paying taxes (we are assuming a closed economy here). How the process of hoarding of money on people’s bank accounts or under their pillows can lead to inflation is beyond my brain’s information processing capacity.

    If people start spending more because they feel wealthy, this is exactly what we wanted to achieve by increasing budget deficit. Yes, excessive spending will lead to inflation. But who said that the transition from hoarding to excessive spending has to be a dramatic avalanche-like process with a positive feedback loop developing? I am afraid that the burden of proof that the transition is to be an avalanche breakdown process is on Prof. Mankiw’s side.

    “Interest rates must then fall to induce people to hold the expanded money supply, again putting upward pressure on aggregate demand and inflation”

    Why would the interest rate need to fall? So that people do not deposit “money” (currency) in banks? But banks do not re-lend deposits. The choice is not between “hoarding money” or “lending it out by depositing in a bank”. The choice is between “not spending” and “spending”. I am afraid that a relapse to Classical Loanable Funds Theory v. 1.0 has just occurred… If people are to be induced to reduce their consumption and companies discouraged from investing, a steep increase of interest rates might help in the short run but this would obviously have side effects (the higher accrual of interests, increasing their Haig-Simons income) so it is pointless to do this. Better to reduce government spending or increase taxes.

    “Third, the increase in inflation reduces the real quantity of money demanded” I would actually agree with this even if I don’t like the term “money demanded”. Households may then initiate capital flight like in Iran or Venezuela. But we are not debating this case here. I believe that CPI needs to be well above 10% to initiate this process or the country needs to be as thoroughly corrupt as Russia where CPI is low but capital flight still occurs. We will need to build a proper SFC model before we can comment further.

  22. Dear Bill,

    Greetings from Greece, I have been following your blog for a couple of years now so I would say that I am rather new to the MMT teachings. So apologies if the following seems to naive or impractical but I was thinking for some time one way to empirically validate the ideas of MMT contrary to the mainstream, which I think is similar to the second point you’ve made on your response and I wanted your opinion on it.
    Since the mainstream statement is that every government spending is offset by debt issuance, wouldn’t it be possible for someone to actually measure the new bonds that a government sells and see if they indeed compensate for the deficit in a specific period of time? If the two are not equal wouldn’t it mean that the mainstream narrative is wrong and MMT is right?

  23. I don’t know if there is some disease that maybe infects economists after a certain amount of time. Maybe like a calcification in their brains or something that causes apparently intelligent people to become unable to understand a series of fairly simple logical arguments no matter how clearly they are made. To the point that they can’t even direct argument against ,if that is what they want to do, in any reasonable direction. Almost all these economists who have written explainers of MMT seem to have that affliction. Lucky for them it doesn’t seem to affect their lifespans in any way. But it does kind of make them seem foolish.

    I have never written a book, let alone an economics textbook, and I do not have a PhD in economics, and have never taught economics, or even pretended to. But I could write a far more accurate “guide” to MMT than what Mankiw produced. And if for some reason I wanted to argue against MMT, I’m pretty sure I could do better at that than these guys also. Makes me wonder sometimes.

  24. Correction on my comment to Dr NRL: that paper was written by the Library of Parliament, a body that provides research and analysis in support of the Canadian Parliament (House and Senate). They have a good reputation for clarity and impartiality on current issues of the day. The BoC is not held in the same regard.

  25. “Greg Mankiw had decided to have no further interaction with us for his own reasons.”

    Wow — has he simply not replied to any follow-ups?

  26. To Bill Mitchell

    Mmt is the only credible alternative to ‘mainstream’ economics both as explanation and prescriptive action.
    The latter will eventually collapse in the weight of its contradictions but it will be a long hard slog.
    Pity the millennials seeking a secure and prosperous future in the current school of Ivy League eco deptts.
    I am not an economist but deeply interested.
    I have a few quick qns.
    You say the ‘central bank does not control the amount of reserves’.
    By reserves, I think you mean banks’ balances with the central bank.
    To my mind, the central bank controls this totally.
    They could buy assets off banks -qe- and add to reserves or sell assets to contract base money.
    ‘Central banks will always supply reserves’.
    I would slightly modify this to state they ‘can’ always supply reserves.
    With this, it makes complete sense that ‘reserves do not constrain lending’.
    If all banks are in ‘liquidity equilibrium’ ( every bank has exactly the reserves, non reserves asset mix that it’s comfortable with),an individual bank will lend only if its risk appetite increases.
    It exchanges reserves for non reserve assets.
    Or borrows from another bank.

Leave a Reply

Your email address will not be published. Required fields are marked *

Back To Top