Bleg: What’s Wrong with the MPC/Spending-Velocity Argument?

I’ve ground the axe quite a bit over the years for the argument that Kevin Drum makes — and dismisses — here.

In brief: poorer people spend a larger share of their income/wealth than richer people. So if poorer people have more income/wealth — if the distribution is more equal — there will be higher money velocity/more spending/more production/higher GDP.

(Search for “marginal propensity” and follow Related Links to see my stabs at this.)

But Kevin — who certainly has the political inclination to make this argument — says:

This sounds pretty plausible, doesn’t it? Higher inequality should generate less consumption, which in turn produces a weaker economy. Unfortunately, the data says something else. “I wish I could sign on to this thesis,” says Paul Krugman, “and I’d be politically very comfortable if I could. But I can’t see how this works.”

Me neither. I spent a couple of months trying to write a magazine piece based on this thesis, and I finally gave up. By the time I was done, I just didn’t believe it. So I gave up and spiked the idea.

I’ve tweeted him and posted a comment, but haven’t heard: what made him give up on this? What convinced him otherwise?

And in response to a recent post where I ground this axe, Scott Sumner responded:

But you really need to give up on that MPC stuff, it was discredited decades ago. Monetary offset rulz.

This in keeping with his seeming assertion that nothing matters except monetary policy, because monetary policy will (or at least should) always offset it.

But still: Sumner, Drum, and Krugman all seem to think that the distribution/MPC/velocity argument has no legs. They’re quite categorical about this.

SRW took a stab at the subject recently, telling a story that I find quite convincing. But didn’t really explain to me why so many feel so certain that it’s not true.

Can folks (especially those who don’t believe this argument) point me to what might be considered definitive takedowns? I have notions about what they might say, but want to see the best argument(s) out there.

These takedowns should, just for instance, convincingly debunk this paper (sorry, gated), which suggests that rising income inequality ’67-’86 resulted in 12% lower consumption spending in ’86 than would have occurred if inequality had remained the same.

Cross-posted at Angry Bear.







32 responses to “Bleg: What’s Wrong with the MPC/Spending-Velocity Argument?”

  1. Arijit Banik Avatar

    What do you think of Michael Kumhof’s work (along with Claire Lebarz, Romain Ranciere, Alexander W. Richter and Nathaniel A. Throckmorton) titled “Income Inequality and Current Account Imbalances” (source:

    It takes a neoclassical approach with DSGE modeling and incorporates Nash bargaining to model the share of income to labour.

    Excerpt (5.3 Increased Inequality)
    “Higher loans from investors increase workers’ leverage, or debt-to-income ratio, from 60 percent to 140 percent after 30 years. This increase, while very large, is still less than what was observed in the data. In the short run, higher debt allows workers to reduce their consumption by less than the drop in their wage, but in the longer run workers’ consumption continues to fall even when their real wage starts to recover (this recovery is due to a slow recovery in bargaining power combined with a rising capital stock).
    The reason is that by this time debt service consumes a far larger and growing portion of workers’ disposable income, around 7 percent compared to 3 percent in the original steady state.”

  2. JKH Avatar

    As always, the correct explanation of saving and MPC depends in part on a correct interpretation of accounting (among other things).

    E.g. SRW says:

    “Household borrowing represents, in a very direct sense, a redistribution of purchasing power from savers to borrowers. So if we worry that oversaving by the rich may lead to an insufficiency of purchases, household borrowing is a natural place to look for a remedy. Sure enough, we find that beginning in the early 1980s, household borrowing began a secular rise that continued until the financial crisis.”

    I don’t know about that.

    Household borrowing occurs mostly in the form of mortgages. To the degree that household borrowing manifested in mortgage borrowing to acquire existing houses, that flow of funds is not reflected in macroeconomic saving. Furthermore, affluent people also took out large mortgages, especially given the tax deductibility of mortgage borrowing in the US. A lot of this activity is within the non-NIPA flow of funds, unrelated to saving rates, and conducted in the land of asset swaps. What that means is that a good part of the correct explanation for macroeconomic saving patterns bypasses a good part of the household credit statistics for the banking system.

    Moreover, new housing demand, which is indeed a NIPA and saving related component, was very much affected by the affluent. And they took out mortgages for this as well.

    What was related to saving and dissaving rates more broadly across different income groups perhaps with a bias to lower income groups was MEW (mortgage equity withdrawal) based spending, and that in turn could be linked fairly directly to the current account deficit. Krugman’s graph on the private sector saving rate correlates somewhat with the deteriorating current account deficit over the same period.

    Anyway, the argument on this issue at the macro level either way requires a counterfactual experiment that cannot be constructed in the laboratory of economics. So the burden of proof lies equally on both sides.

  3. James Oswald Avatar

    This is the central question of macroeconomics. Let’s begin at the beginning.

    Start with an economy with some fixed real productive capacity. At some rate of spending (M*V), prices will be pinned down, because we’ve fixed Y. Now, without any sticky prices or other frictions, allocating more money to poor people will simply cause prices to increase. Prices of things poor people demand will go up, prices of things rich people demand will go down, but because poor people spend money at a higher rate, overall prices will increase.

    Now add in sticky prices. Instead of price increases, you get quantity and price increases proportionate to the slope of the short run aggregate supply curve. Producers face increased demand, so instead of instantly increasing prices and wages, they increase production. In this very simple model, redistribution increases both aggregate demand and total output (as you correctly infer).

    What Scott is saying is that the central bank has a particular target and they can hit that target. The central bank will see the redistribution (whether political or arising from wage changes, or whatever) and reduce the money supply. The reduction in money supply will lower aggregate demand, and cancel out the effect of redistribution.

    Finally, remember that spending on investment adds to aggregate demand just as spending on consumption. Recessions are characterized by big drops in investment and stable consumption, but in theory, you can have stable aggregate demand with more investment and less consumption.

    It’s important to distinguish between long and short runs. Short run you are correct, but there are other actors which are opposing any change (such as the central bank or fiscal policymakers). Long run, changes in demand don’t matter.

  4. James Oswald Avatar

    Inequality changes very slowly relative to prices and central bank reaction speed.

  5. BFWR Avatar

    But what if there is an inherent scarcity of total individual incomes in comparison to total prices, like there actually is by cost accounting convention? And what if this scarcity is constantly enforced on every dollar actually in or even going back through commerce/the economy because cost accounting is such a necessary and deeply embedded part of same, so that velocity theories are basically BS? Cost accounting HAS INDIVIDUAL monetary effects that cannot be overcome by money as LOANS injected into the economy. ONLY a GIFT of money TO THE INDIVIDUAL…can do that.

  6. Asymptosis Avatar

    @James Oswald “The central bank will see the redistribution (whether political or arising from wage changes, or whatever) and reduce the money supply. ”

    Thanks. That’s what I thought. The fed sees some results of the redistribution — wage increases, GDP increase, reduced unemployment, inflation increase, whatever — and it acts less liberally than it would otherwise (the essential counterfactual for this discussion).

    “Short run you are correct, but there are other actors which are opposing any change (such as the central bank or fiscal policymakers). Long run, changes in demand don’t matter.”

    I’m confused here. My impression is that Fed effects are short-term.

    Short-term, yeah, the Fed moves last. But they move in a sea that is not of their making.

    Is this saying that in the long run, nothing matters?

    Two identical 40-year scenarios, one with redistribution, one without. (Assume that the income and substitution effects cancel each other out in all instances, as do all other incentive effects.) There’s just redistribution (or not), and there’s the Fed.

    Is this saying that year-40 GDP would be identical in those two scenarios?

    Isn’t that rather idealized?

  7. James Oswald Avatar

    In the long run nothing matters because prices adjust. Redistribution still matters for other reasons, like politically and just for its own sake, it’s just that it doesn’t matter for the level of spending. In your two scenarios, nominal GDP would be pinned down by the money supply and fiscal policy and real GDP would be determined by the productive capacity of the economy. Price adjustment is the long run great equalizer, but it can take almost a decade.

  8. BFWR Avatar

    The problem is not capitalism versus socialism, its the system versus the individual. Capitalism and socialism have both been given chances in the modern world and both have failed. And it’s BS that “they (either and both) haven’t been tried”. It’s time to evolve a new system by looking closer at the actual workings of the Commerce/the economy. That includes de-constructing all the orthodoxies of the present system because orthodoxies are not always true and/or accurate….as we have come to see. Two of the orthodoxies that need to be looked at are the Quantity theory of money and the velocity of money’s circulation. To that purpose the following theorem:

    Where C/E is Commerce/the Economy, i is an amount less than I, I is total individual incomes, P is total prices and -5 represents the continually enforced scarcity of individual incomes imposed by the realities of cost accounting (labor costs [incomes] are only a fraction of costs and yet all costs must go into price)

    C/E = – 5i + I + P < I + P and – 5i + I + P – I + P = – 5i and cannot equal 0, therefore C/E is not self liquidating.

    Furthermore, if – 5i + I + P persists and/or is a constant systemic condition of total individual incomes because it is caused by an integral part of the system itself (realities of cost accounting), then that system (C/E) is price inflationary and income deflationary.

    The problem cannot be resolved by adding money, especially only money as loans back into the economy, because doing so initiates the system’s scarcity of incomes to prices. Also, any possible or actual money “re-circulating” back through the system re-initiates the same scarcity of individual income. Thus the Quantity theory of money is at best a less than operative factor in inflation and velocity theory is invalid as any re-circulated money re-initiates a scarcity of incomes to prices.

    The solution to the problem is the raising of total incomes by at least 5i via a direct payment/monetary gift to individuals so that no additional cost is incurred, the income scarcity inherent in C/E is avoided/not initiated and thus C/E can be self liquidating.

  9. Greg Avatar

    I see the MPC issue in a different way. Its less about conditions which lead to falling GDP and lower consumption, its about conditions which lead to financial crises…… and THEN falling GDP and consumption because of our improper reaction to the financial crisis

    Too many seem to think that distribution doesnt matter much. As long as total income to the economy is the same demand will be the same, so they seem to argue. I think thats mistaken. I think thats mistaken because it dismisses the toxicity of debt

    Let me offer an analogy to our economy using something I understand well, our cardiovascular system.

    Lets think of the flow of money like the flow of blood through our body. Acid, a normal byproduct of cellular function is like debt. Healthy cells are getting the blood they need and are able to get their waste products, acid, removed. But a cell that is either demanding too much or is in an area that is under perfused, runs the risk of accumulating acids to the point of toxicity and cellular death.

    We have about 6 liters of blood in our body and it gets pumped around at anywhere between 3 liters minute to 15 liters a minute. There is a natural hierarchy of perfusion within the body. The brain, and kidneys alone receive over half of our cardiac output in normal conditions. So while at any given time over half the blood in our bodies is going through brain and kidneys, the rest of the bodies needs must be maintained or their debt (acid) will build up and not just kill that tissue but the whole body. In times when the heart is either incapable to meet the needs (failure) or the needs of the tissue are just too great (infection or vigorous exercise) acids build up. As long as inflows are met, acid wont reach toxic levels but if not sickness or death is imminent.

    Incomes are what keeps consumers healthy. Debts are a normal part of our modern capitalist system as well and as long as the flow of incomes meets debt service their is no problem. But we want growth, and our banks encourage growth by taking on more debt. They want more people taking on more debt. These debts must be cleared or they will reach toxic levels and freeze the system.

    They freeze it in two ways;

    1) They stop getting paid or

    2) They are the ONLY thing getting paid.

    As I said earlier, incomes are what keep consumers healthy and even if total incomes stay close to the same, if that total is only being distributed between 30% less of the consumer force, their debts become toxic and affect the whole system. Banks fail and businesses lose customers.

    When 30%, I imagine its even more than this now, of your potential consumer base is only getting enough income to barely service old debts, these debts hurt everyone. They become a systemic problem not just a personal problem. The rest of the people cannot nor will not increase their debt levels enough to offset this reduction.

    Distribution of flows matter. There are healthy distributions and un healthy distributions

  10. James Oswald Avatar

    @BFWR If you are saying that every human society ever has “failed” maybe your standards are too high. Liberal democracies + capitalism are just about the best thing ever to happen to humanity. Longer life expectancies, higher wealth, better access to food, pretty much any measure of well being ever. BTW, your math is nonsense.

  11. Greg Avatar

    @James Oswald

    Sorry James, we dont live in a Liberal democracy and we really havent had capitalism, weve had centrally planned corporatism.

    Not saying it hasnt raised our living standards, or that it hasnt been mostly good, but it isnt as you describe it. That whole democracy BS has gotta stop. It should be getting embarrassing for those who try and say it with a straight face.

  12. BFWR Avatar

    MY math may be nonsense, but the math of the current system isn’t and the realities which my perhaps flawed mathematical forms point at….are one.

    Technology has raised our living standards…and finance and orthodox economic theory have enslaved us at the same time.

  13. BFWR Avatar


    Your human body analogy is very apt. C. H. Douglas’s Social Credit could be looked at as a philosophically encompassing human body over the present system which brackets the economy from beginning to end with its homeostatic mechanisms of a universal dividend at the inception of the productive process and a compensated retail discount at the end of the productive process.

  14. BFWR Avatar

    Ways Social Credit Dividend improves/transforms economy:
    1) Ends systemically price inflationary and hence individual income and business profit deflationary nature of current system(s) due to flaw in cost accounting rules
    2) In world of advanced technology ends direct negative relationship between efficiency and unemployment
    3) Makes system serve Man instead of other way around
    4) Ends extortionary monopoly power of Banking/Financial system with alternative means of consumer finance
    5) Reveals physical world as abundant instead of scarce
    6) Underpins, energizes and frees money system/society with a psychology of Confidence, Hope, Love and Grace instead of insecurity, hopelessness, mere exchange and scarcity.
    7) Eliminates necessity (and frees ability) to borrow/lend to keep economy “up in the air”, allows it to simply expand (or contract) as individuals so desire
    8) Truly unleashes efficiency of technology instead of sabotaging it like current system
    9) Unavoidably awakens individual to the fact that she/he is responsible for BOTH self AND the system, by ending any possible rationale for them to blame the system for their own lack of economic security
    10) Fits seamlessly within a free market, private property and profit making economy

  15. Eric L Avatar

    There have been some studies showing that, while income equality has grown over the past few decades, consumption inequality has not:

    That to me is pretty strong evidence that debt has been crucial in maintaining demand. Both of those studies ended before the financial crisis. It would be interesting to see what has happened since. The evidence suggests that Sumner *was* right, that income inequality didn’t matter because the Fed could keep lowering rates so that credit would become more widely available to support demand. What he missed is that the Fed has effectively been alleviating inequality by helping the middle class go into debt so they could continue consuming.

  16. Eric L Avatar


    You seem to be getting your idea of how the “affluent” behave by looking at the upper middle class. Yes, people with 6 figure salaries have mortgages too, though they pay them off sooner. But while their incomes have done better than the rest of the middle class, they haven’t taken off in the way that the 7+ figure salary crowd has — and those are the people who own the mortgages that the middle class has taken out.

  17. Eric L Avatar

    @James Oswald

    What Scott is saying is that the central bank has a particular target and they can hit that target. The central bank will see the redistribution (whether political or arising from wage changes, or whatever) and reduce the money supply. The reduction in money supply will lower aggregate demand, and cancel out the effect of redistribution.

    Yes, that is what he believes. I see two problems with it:

    1. Regardless of whether the Fed theoretically can hit their target, there’s a pretty strong case that they haven’t. So if redistribution results in doing what they so far haven’t achieved, there’s no reason they’d undo it out of spite.

    2. If redistribution results in exceeding the Fed’s targets, leading to inflation, they will fight it with higher interest rates. Good. That means rates won’t be so close to zero, which means next time we have a recession, the Fed will be able to respond more effectively.

  18. Greg Avatar

    @Eric L

    “The evidence suggests that Sumner *was* right, that income inequality didn’t matter because the Fed could keep lowering rates so that credit would become more widely available to support demand. What he missed is that the Fed has effectively been alleviating inequality by helping the middle class go into debt so they could continue consuming.”

    Well put

    Trouble with Sumner is he doesnt understand the toxic nature of too much private debt. He thinks those borrowers just owe it to some saver so its a wash. He’s never shown a grasp of how banks work. He models a CB but totally screws up what the private banks are doing and where they fit in to an economy. Just like Nick Rowe who made the statement in one of his recent posts
    that ;

    “Monetary policy does not work by increasing actual borrowing. That is not the causal channel of the monetary policy transmission mechanism. Monetary policy works by increasing spending, not borrowing”

    How do you even start with this nonsense? “Actual borrowing?

    Mike Sankowski at MR had a series of excellent posts about how monetary policies main transmission mechanism was via real estate lending. Rowe and Sumner are so deluded its really kind of scary. They actually have students and readers who are mis-understanding the world in their way, and think they are brilliant. Nick is fond of saying “Macro is hard!!” (I always hear his tone with that statement like the old SNL skit “The Whiner Family” where the mother would say “I’ve got diverticulooooooosis” )

    Well macro is hard when you dont know what banking is about, it would be less hard if you tried to properly model banking into your view.

  19. Greg Avatar


    Social credit sounds interesting. Ill look more at it.

  20. JKH Avatar

    @Eric L

    “those are the people who own the mortgages that the middle class has taken out”

    that is highly dubious, IMO

    the net worth of the top 1 per cent includes a disproportionately high share of equities and direct real estate, and a lower proportionate share of ultimate claims on household debt

    a good chunk of household debt is intermediated directly on the balance sheet of the banking system and another good chunk is intermediated through insurance companies and pension funds; both of those sources include a chunk of GSE liabilities (which intermediate mortgages) as assets

    and the net worth of the top 1 per cent includes a disproportionately low share of those insurance and pension claims and bank deposits

    combine those perspectives, and you find a disproportionately low share of household debt intermediated through to the top 1 per cent

    the top 1 per cent would hold ultimate claims to household debt via the channels mentioned as well as via hedge funds, but all of these would be a disproportionately low system share of overall claims on household debt

  21. JKH Avatar


    and that holds a fortiori for the top .1 per cent and the top .01 per cent

  22. Asymptosis Avatar

    Arijit Banik :

    What do you think of Michael Kumhof’s work (along with Claire Lebarz, Romain Ranciere, Alexander W. Richter and Nathaniel A. Throckmorton) titled “Income Inequality and Current Account Imbalances” (source:

    Just getting to this. Thanks, this is great; it expresses some of what people here have been saying (borrowing to substitute for weak incomes) in terms that neoclassicals might listen to.

    While it skirts near the savers-fund-borrowers cause-and-effect tomfoolery, it attributes causation at a different point in the circle: income inequality causes borrowing which causes current account deficits. Which, I add, reduces domestic real-sector money income (NAFA), in a continuing cycle.

    IOW, it says that income (over which agents have little control) causes saving, not responsible, patient, people.

    Closely related:

  23. BFWR Avatar

    Great. Here are two links for research an a good synopsis:

  24. Greg Avatar


    “IOW, it says that income (over which agents have little control) causes saving, not responsible, patient, people.”


    It has seemed to me, although Ive never seen any of them say it, that income is assumed to be something which we have under our own control. Ive never brought this up before, because it seemed such an absurd assumption, but its something that has nagged me for a long time. No one controls their own income (except Congress I guess since they get to vote for their raises) but owners of capital have many tricks (cutting hours, layoffs) to maintain it even in the face of falling sales. Of course many of these tricks are lauded as “innovations”

  25. James Oswald Avatar

    @Greg I’m perfectly happy defending both democracy in practice and capitalism in practice. Any ideological system is going to hit constraints when you try to implement it in the real world. Obviously the world isn’t perfect, but compared to what?

  26. James Oswald Avatar

    @Eric L
    The central bank makes mistakes, but we should try to figure out what direction they make mistakes in. In normal times, it seems like the err by going above their target, and during financial crises, it seems like they undershoot the target. I don’t know how they are going to err in the future, but that seems to be the story for the last couple of decades. I don’t they’d tighten money out of spite, but they might do it because they couldn’t figure out why prices were increasing. Remember one reason they tightened money in 2008 was an oil shock, which makes no sense in the standard neo-Keynesian model.

  27. Asymptosis Avatar

    Just getting to this. I’d never read Douglas before, will need to think about the cost-accounting implications. (Would love to hear JKH’s thoughts on that.) I’m seeing a lot of connection between your “scarcity of incomes” and Ed Lambert’s thinking about ED and AS: when effective demand (driven largely by labor share of income) drops below AS, things fall apart.

  28. JKH Avatar


    never really looked into it

    but I note the following from Wiki:

    “While John Maynard Keynes referred to Douglas as a “private, perhaps, but not a major in the brave army of heretics, he did state that Douglas “is entitled to claim, as against some of his orthodox adversaries, that he at least has not been wholly oblivious of the outstanding problem of our economic system.”… Keynes said that Douglas’s A+B theorem “includes much mere mystification”

    Keynes did understand accounting, so that tempers my curiosity somewhat

  29. Socred Avatar


    Keynes did understand accounting, but his philosophical approach to the problem, due to the fact that he had socialist tendencies, led to a different solution. Further, he failed to see the underlying cause of the deficiency between income and prices because he was solely focused on the deficiency between income and the price of consumer goods (which is often temporary), whereas Douglas focused on the deficiency between income and the total price of goods (consumer goods + capital goods). Keynes thought that the deficiency between total incomes and the price of consumer goods could be overcome by increasing the production of capital goods and thus increasing incomes paid to the workers who construct the capital. But even Keynes recognized the folly of this “fix” when he wrote:

    “Thus the problem of providing that new capital-investment shall always outrun capital-disinvestment sufficiently to fill the gap between net income and consumption, presents a problem which is increasingly difficult as capital increases. New capital-investment can only take place in excess of current capital-disinvestment if future expenditure on consumption is expected to increase. Each time we secure to-day’s equilibrium by increased investment we are aggravating the difficulty of securing equilibrium to-morrow.” (The General Theory of Employment, Interest and Money)

    This is because the cost of said capital eventually makes its way to the cost of consumer goods, which just makes the attempt to equate income and the price of consumer goods more difficult at a future point in time.

    Douglas realized this exact same thing over a decade earlier when he wrote:

    “In the first place, these capital goods have to be sold to someone. They form a reservoir of forced exports. They must, as intermediate products, enter somehow into the price of subsequent ultimate products and they produce a position of most unstable equilibrium, since the life of capital goods is in general longer than that of consumable goods, or ultimate products, and yet in order to meet the requirements for money to buy the consumable goods, the rate of production of capital goods must be continuously increased”

    In other words, Keynesianism is not a permanent solution to the problem, only Douglas’s proposal for a national dividend and compensated price are a permanent solution.

    Douglas testified before the MacMillan Committee on Finance and Industry, in which J.M. Keynes questioned him.

    The transcript can be found at the link below:

  30. BFWR Avatar

    Thanks for that accurate and more comprehensive look at both Social Credit and Keynesianism.
    Theories are ABOUT reality hopefully. Douglas’s theories and A + B theorem were taken directly FROM empirical data FROM one of the most deeply embedded tools and necessities of commerce/the economy itself. In other words they are built DIRECTLY on commercial REALITIES, not abstraction formulated ABOUT them. I like the idea of realities trumping theories. I hope your input and also the one I just made pique the curiosity of the posters

  31. James Oswald Avatar

    I’m sorry, but I can’t make heads of tails of your comment.