How many economists can answer this question?

Dustin asks an interesting question:

An elementary question on the topic of interest rates that I’ve been unable to resolve via google:

Regarding Fed actions, I understand that reduced interest rates are thought to be expansionary because the resulting decrease in cost of capital induces greater investment. But I also understand that reduced interest rates are thought to be contractionary because the resulting decrease in opportunity cost of holding money increases demand for money.

One? Neither? Both? Little of each? Depends?

It’s not at all clear that lower interest rates boost investment (never reason from a price change.)  And even if they did boost investment it is not at all clear that they would boost GDP.

But two things are very clear:

1.  Open market purchases reduce short term nominal rates.

2.  Open market purchases boost NGDP.

However it’s surprisingly hard to explain why OMPs boost NGDP using the mechanism of interest rates. Dustin is right that lower interest rates increase the demand for money.  They also reduce velocity. Higher money demand and lower velocity will, ceteris paribus, reduce NGDP.  So why does everyone think that a cut in interest rates increases NGDP?  Is it possible that Steve Williamson is right after all?

Here’s what people forget.  The Fed doesn’t wave a magic wand and reduce interest rates.  They do so via a boost in the monetary base.  Now let’s think about the effect of a sudden and instantaneous 1% boost in the monetary base. Contrary to most macro models, there is an immediate impact on NGDP, although almost certainly less than 1%.  The instantaneous impact on NGDP comes from the fact that the OMP immediately boosts commodity prices, and hence the gold and oil flowing out of the ground in America has a higher nominal value.  But that’s nowhere near enough to instantly boost NGDP by 1%.  Instead interest rates must fall by enough so that Americans are willing to hold extra non-interest bearing base money.  The lower interest rates reduce velocity in the short run.  So NGDP might rise by 0.2%, and V might fall by 0.8%, within seconds.

In the very long run money is neutral, V is unchanged, and NGDP rises by 1%.

The medium term is the most complicated.  There may well be a period when velocity goes up, especially if inflation rises and real growth is strong.

So the correct answer is that the lower interest rates tend to reduce NGDP, but the thing that causes lower interest rates may increase NGDP by more than the reduction caused by lower V.  At least if that “thing” is OMPs.

Yes, easy money often makes rates fall in the short run, but it’s the larger money supply that does the “heavy lifting” of boosting NGDP.

And here’s my claim, if a student asked Dustin’s question in a typical American econ class, the professor would struggle to answer the question.  What do you think?  Note that to answer the question you must address the specific concern raised by the student (lower rates causing rising demand for money in this case.) It’s not enough to wave off that concern and answer it using an unrelated model, say New Keynesianism.

Now suppose interest rates were cut via lower interest on reserves. How would the answer be different? Lower IOR would reduce the demand for money, which would boost NGDP.  No increase in the monetary base would be necessary.  No wonder that many anti-monetarists are so anxious to move to a world of IOR. They are embarrassed by some inconvenient facts about the impact of lower interest rates in a world of OMOs involving zero interest base money.

PS.  Some have claimed that in the brave new world of IOR we won’t need OMPs to target NGDP.  Perhaps with a zero growth rate target path that would work. But if you want 5% NGDP growth, then IOR would have to fall fast enough to lead to 5% velocity growth.  Eventually you would blow right through the zero bound, and need negative IOR.  Even on currency.  Velocity would rise faster and faster. After being paid workers would run to the store to spend their money.  Then they’d start using jetpacks.  Eventually the speed of light might put an upper limit on V, and hence NGDP.

I’d rather stick to good old OMPs.



44 Responses to “How many economists can answer this question?”

  1. Gravatar of Vaidas Urba Vaidas Urba
    23. December 2013 at 10:40

    The question is – how do we assign instruments to targets when we have both IOR and OMOs? We assign IOR to NGDP, and we assign OMOs to complying with the Friedman Rule.

  2. Gravatar of Kenneth Duda Kenneth Duda
    23. December 2013 at 12:26

    Scott: I have a question for you. You wrote:

    > The instantaneous impact on NGDP comes from the
    > fact that the OMP immediately boosts commodity
    > prices, and hence the gold and oil flowing out of
    > the ground in America has a higher nominal value.

    I get it if the 1% increase in M is believed to be permanent. But what if the markets think the Fed is mostly just kidding, that as soon as there is any whiff of inflation, the Fed is going to snatch the punchbowl away and sell its assets and stick M right back down where it was? Given those expectations, it seems to me a 1% increase in M will have no effect on commodity prices at all.

    Today’s Fed lacks credibility to be irresponsible, and is thus finding it difficult to increase the price level even in the medium term. In the long term, of course money is neutral, but the long term may end up being really long indeed if the markets believe the Fed will reduce M as soon as the price level starts to increase.

    Which is, of course, why we need NGDPLT.


    Kenneth Duda
    Menlo Park, CA

  3. Gravatar of ssumner ssumner
    23. December 2013 at 13:19

    Vaidas, I’m not a fan of the Friedman rule. I’d prefer positive interest rates. Or did you mean use IOR to apply the Friedman rule? That would not cover cash.

    And I’m not a fan of IOR in general. I’d prefer to get rid of reserve requirements and rely totally on OMOs (perhaps with occasional discount loans, although I wonder if they are needed either.)

    I’d prefer they use NGDP futrures as a policy instrument, not IOR.

    Kenneth, I don’t see any evidence at all that the Fed lacks credibility. Rather they have the wrong target. If they had the right target, things would look much better right now. Whenever the Fed changes policy markets react strongly, which suggests they have plenty of credibility.

    If they lacked credibility and really did want higher inflation then they certainly would not have tapered last week. They tapered because they felt they had achieved their goals.

    To answer your question, if the injection is temporary the size of the commodity price response depends on how temporary.

  4. Gravatar of OhMy OhMy
    23. December 2013 at 13:29

    So by your calculation when Bernanke doubled the base the NGDP “instantaneously” increased in December 2008 by 0.2*100%=20%. Interesting.

  5. Gravatar of Mark A. Sadowski Mark A. Sadowski
    23. December 2013 at 13:48

    “It’s not at all clear that lower interest rates boost investment (never reason from a price change.)”

    I ran a whole series of Granger causality tests on Mishkin’s elementary nine channel Monetary Transmission Mechanism this year and a lot of the results were counter to conventional wisdom but were not really that surprising to me personally (and probably not to you either). My sample period was 1993Q1 through 2013Q2.

    With respect to the Traditional Real Interest Rate Channel, I found that the real (adjusted by the year on year PCEPI) 10-year T-Note yield Granger causes private nonresidential investment (PNFI) and private residential investment (PRFI) but I found no correlation with durable goods spending. More importantly, the impulse response is *positive* in each case. In other words *higher* real interest rates lead to higher investment spending. That only makes sense if you know that interest rates are an epiphenomenon.

    P.S. I also checked causality between the real 10-year T-Note yield and the fed funds rate and found that the real 10-year T-Note yield Granger causes the fed funds rate (but not the other way around) and that the impulse response is positive. The endogenous money people will of course interpret that as meaning the wind and current are controling the rudder.

  6. Gravatar of Jaap de Vries Jaap de Vries
    23. December 2013 at 15:51

    Maybe I am missing a point, but usually the communication channel makes sure interest rates adjust when the Fed makes an announcement? They usually do not actually need to go in the market and do OMPs? How does this change the analysis?

  7. Gravatar of benjamin cole benjamin cole
    23. December 2013 at 16:32

    Okay, here is my question: If the Fed monetizes federal debt (through OMP), at what point does it become inflationary?
    At $1?
    No, too small.
    Then $10 million?
    Too small.
    You see my point—at some point OMP must start becoming stimulative and then inflationary. That’s where we should be now. As for “runaway” inflation, the OMP can always be reversed.

  8. Gravatar of OhMy OhMy
    23. December 2013 at 16:35

    Jaap de Vries,

    It changes the analysis in the sense that it exposes it to be a total and utter nonsense. Scott Sumner seems unaware that rate changes have little to do with monetary base, it is mostly done by announcement.

  9. Gravatar of Vaidas Urba Vaidas Urba
    23. December 2013 at 16:55

    Scott, these days cash is ignored, and Friedman rule is achieved in the context of IOR.
    Most of the time, monetary policy works well without IOR, and distortions corrected by the Friedman rule are not important. The only time Friedman rule is really relevant is during financial panic.

  10. Gravatar of Dustin Dustin
    23. December 2013 at 17:43

    Brilliant, Scott!
    Unfortunately reality doesn’t seem to allow for an entirely simple answer. At least I do need to think about rates differently.

  11. Gravatar of Geoff Geoff
    23. December 2013 at 19:04

    This question depends on how one defines expansionary and contractionary.

    If one defines them in terms of the aggregate money supply, then we can conclude that money was highly contractionary in the early 1930s, early 1990s, and late 2000s, when the rate of money growth turned negative; and money was highly expansionary during the 1970s and the mid-2000s, when the rate of money growth exceeded 12% per annum.

    NGDP is not a reliable statistic, because it fails to account for the demand for money holding as an integral part of economic planning, coordination, and forecasting.

  12. Gravatar of Don Geddis Don Geddis
    23. December 2013 at 21:30

    Scott: A student asks a question about the effects of Fed actions, and even your answer doesn’t mention expectations? Aren’t you both stuck on the concrete steppes?

    Surely, by far the most dominant overall effects, come from Fed communication and thus public expectations about the future. Isn’t your answer a bit like answering a student’s question about the fiscal multiplier, without considering the monetary policy reaction function?

    I would suggest instead that: sure, greater investment, and sure, increased demand for holding cash … but both are dominated by whatever change in the future path of NGDP is now newly expected, because of the Fed’s surprising new action.

  13. Gravatar of Geoff Geoff
    23. December 2013 at 22:04

    “But if you want 5% NGDP growth,”

    According to the principles of market monetarism, any constant growth rate is as good as any other, as long as it is constant and credible.

    There is no basis for “wanting” one rate over another that is founded on market monetarist principles.


    “I would suggest instead that: sure, greater investment, and sure, increased demand for holding cash … but both are dominated by whatever change in the future path of NGDP is now newly expected”

    Not even close. Everyone but market monetarists do not care a single iota about “NGDP”. No business uses it when making investment decisions. No investor uses it when choosing what prices to bid and ask. The market has not created NGDP futures.

    NGDP is a useless statistic, if we assume the market is the most correct.

  14. Gravatar of TravisV TravisV
    23. December 2013 at 22:46

    Prof. Sumner,

    Is China’s central bank intentionally tightening in order to “prick” its real estate bubble?

    How much more can they possibly tighten until political pressure forces them to stop?

    Another thought: how does cheap oil factor in to the central bank’s analysis?

  15. Gravatar of Ralph Musgrave Ralph Musgrave
    24. December 2013 at 00:22

    Good to see monetarists in a complete muddle as to whether interest rate adjustments or OMP raises demand and if so why. Why not just abandon the whole idea? Here’s a better way of raising demand.

    Print new money and raise public spending. Second, boost household incomes e.g. via tax cuts and fund that with new money. Households (amazingly) spend more when their income and stocks of cash rises. What do households do when they win a lottery or get a tax rebates? No prizes for the right answer because the answer is so simple.

  16. Gravatar of James in London James in London
    24. December 2013 at 01:51

    Some households may spend more after tax cuts, but to know what most do you have to examine their expectations. For a proper study, see this:

    What’s your proper academic evidence?

  17. Gravatar of Dustin Dustin
    24. December 2013 at 06:53

    It’s not at all clear that lower interest rates boost investment (never reason from a price change.) And even if they did boost investment it is not at all clear that they would boost GDP.”

    I suspect it would be a tenuous claim that active investment flows increase due to reduced rates, though nominal investment? Simply the appreciation of current holdings’ nominal values? That would seem to be sufficiently expansionary as it offsets the underwater effect of nominal debt as well as making me feel better about my 401k…

    This would seem to be powerful in the short run to buffet expectations of near term growth (and thereby lead to growth today), whereas the contractionary angle of increased money demand may be slow and winding. Otherwise, I can’t make sense of why a rate reduction is even useful as I have always just taken for granted that it is!

  18. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    24. December 2013 at 08:28

    ‘Good to see monetarists in a complete muddle as to whether interest rate adjustments or OMP raises demand….’

    Demand for what, money or goods and services?

  19. Gravatar of Don Geddis Don Geddis
    24. December 2013 at 09:23

    Ralph Musgrave: “a complete muddle as to whether interest rate adjustments or OMP raises demand

    OMPs raise demand via the HPE. The only “muddle” is for people who think that the Fed raises demand by lowering interest rates. That transmission mechanism is less clear.

    There is no confusion about whether OMPs raise demand. They do.

  20. Gravatar of Brian Donohue Brian Donohue
    24. December 2013 at 09:46


    I have 17,258,331,500,000 reasons to object to your recommendation.

  21. Gravatar of ssumner ssumner
    24. December 2013 at 10:52

    OhMy, I see your reading comprehension skills have not improved. I said exactly the opposite, a change in the base results in little immediate impact on GDP, even if permanent (which the recent increases have not been.)

    Mark, Good point. Perhaps the markets figure out when rate cuts are needed before the Fed does. (No surprise.)

    Jaap, Yes rates often move in anticipation, but the rate won’t stick unless the base is adjusted to equate supply and demand. Also remember that many rate cuts are in response to a weakening economy—in which case no extra money is needed. BTW, the base changes every day.

    Vaidas, In 2007 cash was 99% of the base, not sure why it is “ignored.”

    Don, I don’t see how any of that changes my answer, which did incorporate expectations, contrary to your claim. Both the response of interest rates and commodity prices occurs within seconds. How does that not reflect expectations? What part of my answer do you think is wrong? (Note, this is not a market monetarist view here, it’s really the only way to answer the question. I can’t imagine a Keynesian disagreeing with anything I wrote here. They aren’t going to claim 1% more money makes NGDP rise instantly by 1%. So V must fall in the short run. I’m not doing anything controversial here.)

    TravisV, There is no real estate bubble in China, and there won’t have been one even if prices fall 25% in the next year. I have no idea what the PBoC is up to.

    Ralph, What muddle?

  22. Gravatar of Scott H. Scott H.
    24. December 2013 at 11:09

    A physics analogy my be useful here. When we want to calculate the resulting acceleration of a particular object, we must first calculate the sum total of forces acting on that object. Only then can we gather the direction and magnitude of the acceleration (assuming we know the mass).

    The situation is very similar with this type of economics question. There are many different forces and many different starting conditions that will influence the direction of the economy. It seems counter productive to try and make blanket statements about outcomes to policies if the relevant starting conditions and forces are not well understood or even established a priori as being relevant. There can easily be sets of forces or conditions that make generalities, while generally true, not true in any particular case.

    For that matter using English instead of math makes things more difficult. A “gotcha” at the end of every blog entry doesn’t contribute either.

    Can you imagine this in physics… “First he said the sum total of the forces would accelerate the object southwards. Now, just 3 months later, he says the sum total would accelerate the object northwards. Why can’t this guy just make up his mind?”

  23. Gravatar of Vaidas Urba Vaidas Urba
    24. December 2013 at 13:29

    “In 2007 cash was 99% of the base, not sure why it is “ignored.””

    In 2008 demand shock for reserves was much larger than demand shock for cash. This is the primary reason why Friedman rule is discussed for reserves these days. Nobody wants to implement Friedman rule for cash these days.

  24. Gravatar of ssumner ssumner
    25. December 2013 at 05:49

    Scott, Good point.

    Vaidas, Why not just accommodate the extra demand for reserves via a OMP?

  25. Gravatar of Vaidas Urba Vaidas Urba
    25. December 2013 at 08:42

    Yes, extra demand for reserves should be accommodated via OMPs. Friedman rule says that not paying interest on reserves is equivalent to taxing them, so we are comparing two policies – fully accommodate demand by issuing interest bearing reserves, or you can issue less of them by taxing reserves. The first policy is a better one, as the second one is equivalent to a tax on financial sector that is getting larger as financial panic deepens.

  26. Gravatar of John Ward John Ward
    25. December 2013 at 12:30

    Please don’t think me rude, but I think that trying to repair what’s irreparable is complicating a simple but deadly problem.
    Keynes’s view on stimulation was “it’s not a tickling contest”: you try it bigtime for a predetermined time, if it doesn’t work, look elsewhere.
    The Fed and the BoE have been trying it up down and sideways for nearly five years, and it has had (econometrically speaking) no discernible effect. Equally, Zirp hasn’t done anything beyond make life easier for the banks. QE has made life easier for the stock market. That’s it.
    The smart and algorithmed money has used both tactics in the worst possible sense of the verb. They now know the Party’s over.
    Globalism as a theory (in the mercantile sense) is self-destructive, neoliberalism is as inhumanely flawed as command socialism, investment banking has lost the plot re primary function, and the credit-growth paradigm has run out of road.
    Forget all the acronyms, and the Ivy League-meets-Wall-Street faux science: we will not get back to economies based on fundamentals until everyone abandons old ideas in favour of new goals. If there is disagreement on the goals, then how on earth can we create news strategies?
    At times in history, radicalism is responsibility. This is one of them. The sooner we start, the less painful the process will be…and we are already nine years behind schedule.

  27. Gravatar of Doug M Doug M
    26. December 2013 at 10:16

    Which rates?
    Most of the time the Fed targets one rate, the rate at which banks lend to one another, which has very little to do with the rate at a person can borrow money from a bank, or the expected rates of return of financial assets, or the opportunity cost for holding cash.

  28. Gravatar of Fed Up Fed Up
    27. December 2013 at 23:00

    Dustin, define “money”.

  29. Gravatar of Fed Up Fed Up
    27. December 2013 at 23:54

    “Here’s what people forget. The Fed doesn’t wave a magic wand and reduce interest rates. They do so via a boost in the monetary base.”

    1) I’m going to assume by “interest rates” you actually mean the fed funds rate. Currency yields 0%. IOR is 0%.

    2) The fed can lower the fed funds rate without increasing the monetary base. Fed funds rate is 4%. The fed announces the new fed funds rate target is 2%. The fed can buy gov’t bonds/sell reserves to the banking system. There are excess reserves in the banking system. The fed funds rate starts falling towards zero. The fed sells the gov’t bonds/buys the excess reserves from the banking system at 2%.

    Currency yields 0%. IOR is 0%. Monetary base is unchanged. Fed funds rate goes down to 2% from 4%.

  30. Gravatar of Dustin Dustin
    28. December 2013 at 00:15

    Scott H:
    In this case the question is regarding a single force that has apparently counterveiling forces. I do not know of an example in physics where a single force opposes itself, singularly producing northern and southern movements simultaneously. Gravity doesn’t repel AND attract.

    Doug M:
    Good question! My original question was inspired by those rates that the Fed strictly controls… Ie, short term rates. Though upon reading Scott Sumner’s response I realized i made an error in conflating the long term and the short term.

    Fed Up:
    Money would be the monetary base

  31. Gravatar of Ralph Musgrave Ralph Musgrave
    28. December 2013 at 00:38

    James in London,

    The study you refer to says “We find that while almost all of the tax and benefit changes examined in the study prompted changes in consumer spending…” So that supports my point. Other studies which (amazingly) find a relationship between household income increases and household spending are thus:

    Of course I’m not saying that the latter relationship is simple or entirely predictable. But given the complete muddle over the effect of interest rate adjustments and QE, tax cuts don’t fare too badly.

  32. Gravatar of Ralph Musgrave Ralph Musgrave
    28. December 2013 at 00:49

    Brian Donohue,

    What’s the big problem with the debt? The real rate of interest paid on it is about zero (or even negative). As to the possibility that interest demanded for holding US debt may rise, that has no effect whatever on the interest paid in EXISTING DEBT. I.e. it’s only NEW (e.g. rolled over) debt that’s affected. But there’s an easy solution to that problem: DON’T BORROW!!! I.e. print instead.

    I mean what on Earth is the point of an entity that can print money borrowing the stuff? (I’m referring to the US government / Fed). As to the AMOUNT that needs to be printed, that needs to be enough to bring the economy up to capacity, but not so much as to cause excess inflation.

    I fail to see the problem.

  33. Gravatar of Geoff Geoff
    28. December 2013 at 01:13

    “I mean what on Earth is the point of an entity that can print money borrowing the stuff?”

    Because it’s less inflationary obviously.

  34. Gravatar of flow5 flow5
    28. December 2013 at 09:54

    “Open market purchases reduce short term nominal rates”

    Open market operations of the buying type only decrease or suppress short-term interest rates whenever the rate-of-change in money flows isn’t on the upswing (the 24 month proxy for inflation). This is largely a reflection of inflation expectations (the expectation that price level will chronically increase injects an “inflation premium” into the yield curve).

    Paul Volcker’s experiment (using Paul Meek’s antiquated 1974 operating procedures), represents a spectacular example of this phenomenon.

    But the Fed’s “open market power” has been emasculated. QE or POMOs (under the payment of interest on excess reserve balances), resulted in, at the very most, only 39 cents of every dollar of assets purchased by the FRB_NY’s “trading desk” being converted into new money.

    Prior to Oct 3, 2008, the expansion coefficient was 208 dollars of loans and investments for every dollar of assets purchased by the FRB_NY (the multiplier equals the base, i.e., required reserves, divided by commercial bank credit). To say the remuneration rate is a credit control device is an understatement.


    Quantitative Easing and Money Growth:
    Potential for Higher Inflation?
    Senior Economist: Daniel L. Thornton FRB_STL

  35. Gravatar of Fed Up Fed Up
    29. December 2013 at 12:24

    Dustin said: “Money would be the monetary base”


  36. Gravatar of flow5 flow5
    30. December 2013 at 13:57

    Fed Up:

    The answer is: Money is the measure of liquidity.

    Dr. Leland Pritchard: If the “store of purchasing power” attribute of money, when applied to a given asset, is to have significant meaning, it ought to be defined in terms which are applicable to the whole economy. That is, no asset really has a “monetary store of purchasing power” quality unless there can be a net conversion of that asset into money, ceteris peribus. In other words it must be possible to effect this conversion without necessitating that any present money holder reduce/liquidate his holdings/assets. Any other interpretation becomes mired in a futile discussion of relative degrees of confidence and liquidity. But much more than monetary liquidity for the individual holder is necessary if an asset can be said to have the “store of purchasing power” quality; it must be simultaneously monetarily liquid for society as a whole.

  37. Gravatar of flow5 flow5
    30. December 2013 at 14:06

    Money (less Vt):

    2013-01 ,,,,, 0.169
    2013-02 ,,,,, 0.154
    2013-03 ,,,,, 0.156
    2013-04 ,,,,, 0.15
    2013-05 ,,,,, 0.136
    2013-06 ,,,,, 0.073
    2013-07 ,,,,, 0.088
    2013-08 ,,,,, 0.072
    2013-09 ,,,,, 0.075
    2013-10 ,,,,, 0.018
    2013-11 ,,,,, 0.068
    2013-12 ,,,,, 0.092 rebound’s over early unless Vt rises
    2014-01 ,,,,, 0.066
    2014-02 ,,,,, 0.041
    2014-03 ,,,,, 0.065

  38. Gravatar of Geoff Geoff
    30. December 2013 at 14:34

    Fed Up:

    “…That is, no asset really has a “monetary store of purchasing power” quality unless there can be a net conversion of that asset into money, ceteris peribus. In other words it must be possible to effect this conversion without necessitating that any present money holder reduce/liquidate his holdings/assets.”

    This is impossible. It is impossible for A to acquire money from B, without B giving money to A, which is to say B reducing their balance of money, ceteris paribus.

  39. Gravatar of flow5 flow5
    30. December 2013 at 16:09


    You’ve learned your catechisms – but not much of anything else.

  40. Gravatar of Geoff Geoff
    30. December 2013 at 16:12


    I understand that you’re afraid to engage my arguments.

  41. Gravatar of ssumner ssumner
    31. December 2013 at 11:20

    Fed up, When the Fed targets the fed funds rate the base is endogenous. They can only cut the fed funds target without boosting the base when the demand for base money shifts to the left at the same moment.

  42. Gravatar of flow5 flow5
    31. December 2013 at 14:05


    No, goes back to when my best friend persuaded my science teacher to switch lab partners with the best looking girl in the school. She complained to my teacher that I wouldn’t explain my work.

    I still haven’t finish reading your comment. But “breaking the buck” comes to mind. So does the source of time deposits to the CB system. Play with someone who’s time’s less valuable.

  43. Gravatar of Geoff Geoff
    1. January 2014 at 19:44


    Got it. Afraid.

  44. Gravatar of flow5 flow5
    6. January 2014 at 10:43

    NO. Sorry I finally read it. A never gave up the money. Translating microeconomics into macroeconomic principles seems impossible for the layman.

Leave a Reply