Fiscal and monetary policy are not alternatives
I am constantly running into commenters who seem to think that fiscal and monetary stimulus are alternative policies. They are not, unless you envision going to barter. So let’s try to figure out what people like Larry Summers are actually advocating.
I favor setting monetary policy at a position where AD is expected to grow at a socially desirable rate. Suppose someone says; “No, that’s a bad idea, we should use fiscal policy.” And suppose he gets free rein over fiscal policy. And suppose it works, i.e. that AD grows at a socially desirable rate. What then? Then I’d say you’ve just adopted my preferred monetary policy—you set monetary policy at a position where AD was expected to grow at the socially desirable rate.
Fiscal policy can’t really do anything in the AD/NGDP area. So what do people like Larry Summers mean when they talk about a preference for using fiscal policy? They aren’t advocating the use of fiscal policy to get the right level of NGDP growth; you can do that with monetary policy. They are not recommending that fiscal policy be used to attack unemployment, they are recommending that fiscal policy be used to attack the private sector. And that’s because they believe that when interest rates are low the private sector is not efficient, at least compared to the public sector.
And by the way, this is not my mischievous interpretation of the Summers view, both Summers and his follows are quite open in stating that this is the real reason for their advocacy of fiscal policy.
So please don’t waste my time with silly arguments about the advantage of fiscal policy over monetary policy, unless you are advocating barter. Say you want fiscal policy because you think the economy needs more socialism and less capitalism. That a perfectly respectable argument, so make it. Don’t beat around the bush.
PS. And don’t call it fiscal stimulus either. Tax cuts are fiscal stimulus, and as you might have noticed almost all the Keynesians favored the tax increases Obama adopted a year ago. You don’t favor fiscal stimulus; you favor more government spending. And not transfers, those are tax cuts too. You favor more government OUTPUT. You’ve rejected neoliberalism, so say so.
Tags:
14. January 2014 at 17:02
>>they are recommending that fiscal policy be used to attack the private sector.>>
Where does Summers say this?
14. January 2014 at 17:32
foosion, He says the private sector will spend money on wasteful investments when rates are low, so it would be better for the government to spend the money. DeLong has a good post explaining what this is all about.
14. January 2014 at 17:36
Scott, you said
“What then? Then I’d say you’ve just adopted my preferred monetary policy””you set monetary policy at a position where AD was expected to grow at the socially desirable rate.”
Isn’t this just question begging, on its own? I’m sure you have a clearer justification for it that you’ve written before, but on its own in this post it just seems like assuming your conclusion. Maybe you could link to another post on this, or add a bit of exposition, to clarify that you mean that nominal income growth is determined by the central bank’s actions, whereas the government’s actions can only alter its real composition (holding central bank action “constant”).
14. January 2014 at 17:46
Excellent blogging.
BTW, on the federal level, by far the largest spending (financed by income taxes) takes place at DoD, DHS, and the VA. Those three agencies eat up $1 trillion a year. The rest of the federal agency government is almost an afterthought.
This would be much more clear if we use the old federal budget, pre LBJ’s “unified” budget.
Take out Social Security and Medicare, and you see a federal government devoted to national security and rural subsidization…
14. January 2014 at 17:46
Scott,
I don’t understand how these statements are not contradictory:
“And suppose he gets free rein over fiscal policy. And suppose it works, i.e. that AD grows at a socially desirable rate.”
“Fiscal policy can’t really do anything in the AD/NGDP area.”
Either fiscal policy didn’t do anything in the first statement, and AD miraculously grew at a socially desirable rate, or fiscal policy directly affected AD. What am i missing?
14. January 2014 at 17:54
I sort of agree with this, except replace “fiscal policy” with “increased government spending.” I think all changes to government spending decisions should be decided on a cost/benefit analysis. But macro conditions should play a big role in taxes and transfers.
Another form of fiscal stimulus isn’t so much about increasing government spending, it’s about preventing cuts by providing aid to states. Just as spending increases should be made on a micro benefit/cost basis, spending cuts should be subjected to that kind of analysis too, rather than being based on the arbitrary fluctuations in government revenues due to the business cycle.
14. January 2014 at 18:08
slowclap.gif
14. January 2014 at 18:30
Maybe they think the government can better get around the governments own regulatory hurdles (the major reason for private sector inefficiency)..but this too is an unrealistic stretch 🙂
14. January 2014 at 18:46
Scott,
Perhaps the best statement you’ve ever made. And pithy!
14. January 2014 at 18:52
So if Larry Summers argues for fiscal policy to grow the size of government you advocate monetary policy to shrink its size. It sounds like an alternative to me.
Here I think you almost have a Eureka moment. You accuse Summers of advancing fiscal policy for a political agenda. All that’s left is to admit that you too have an agenda and that’s why you argue for monetary policy.
14. January 2014 at 18:54
I think this assumes that the effect of monetary policy on, say, the price level and the effect of fiscal policy on it are in the same one-dimensional space.
I have some pictures to show this may be a higher dimensional problem (monetary and fiscal policy are separate dimensions and therefore can be “alternatives”):
http://informationtransfereconomics.blogspot.com/2014/01/three-dimensional-thinking.html
14. January 2014 at 18:56
That’s really always been my point. Some economists like to poise as doing some kind of pure science innocent of any sordid political agenda. In fact there’s always a political dimension to economics.
14. January 2014 at 19:07
Scott,
Your second paragraph seems completely incoherent to me. Suppose I’m given free rein over fiscal policy, and implement a (Larry) Summers-style fiscal policy consisting entirely of debt-financed infrastructure spending. Suppose that NGDP growth then increases to 5%. So now I’ve just implemented your monetary policy??? WTF???
I’ve been reading your blog for several years now, and I honestly want to understand market monetarism. But paraphrasing a former colleague of mine, this post subtracts value.
PS: I have a PhD in economics and am not related to Larry.
PPS: I’m guessing you’d say that the hypothetical Fed didn’t offset my hypothetical fiscal policy, right? A fundamental question I’ve had all along is how if dY/dG = 4 and dY/dM = -4, then dY/dG + dY/dM = 0 proves dY/dG = 0. I’m probably missing something, but what?
14. January 2014 at 19:23
Mike Sax said:
“So if Larry Summers argues for fiscal policy to grow the size of government you advocate monetary policy to shrink its size. It sounds like an alternative to me.
Here I think you almost have a Eureka moment. You accuse Summers of advancing fiscal policy for a political agenda. All that’s left is to admit that you too have an agenda and that’s why you argue for monetary policy.”
It seems to me that one could implement NGDPLT while still setting government spending at whatever % of NGDP is desired.
In fact, I’ll do it right now. I propose that the US government increase the tax burden to 48.2% of GDP and increase government spending to 56% of GDP. That would put us right where Denmark is, a country that Prof. Sumner has written highly of several times on this blog.
Such as here: http://www.themoneyillusion.com/?p=368
14. January 2014 at 20:05
“And that’s because they believe that when interest rates are low the private sector is not efficient, at least compared to the public sector.”
Unk. I never noticed until now that the liquidity trap model succeeds by failing — interest rates never rise, so you need more fiscal spending… and more… and more…
This is why conservatives should hew to market monetarism.
14. January 2014 at 20:10
Mike, You said;
“That’s really always been my point. Some economists like to poise as doing some kind of pure science innocent of any sordid political agenda. In fact there’s always a political dimension to economics.”
If I was a socialist my views on monetary policy would be exactly the same, target NGDP. I guess you never heard about the liberals who favor NGDPLT.
Garrett gets it, so why can’t you?
Everyone, If my monetary policy is a 5% NGDP growth target, then anything the government does to create 5% expected NGDP growth effectively means they’ve implemented my monetary policy. Now they may have also done other things, and I may or many not approve of those other things, but they’ve implemented my monetary policy.
MM is not about interest rates or the monetary base, it’s about expected NGDP growth. Get that on target and you’ve got a MM monetary policy. It doesn’t matter whether you use QE, fiscal stimulus, currency depreciation or or a magic wand. Of course if you do fiscal stimulus you’ve done more than implement a MM monetary policy. But if you are successful you have at least done that. If you do it by eliminating the corporate income tax (something I favor) I’ll applaud. But I’ll still see the tax cut as being essentially unrelated to monetary policy. If it causes the Fed to “see the light,” that’s great. But in the end of the day it’s up to the Fed.
14. January 2014 at 20:34
Fiscal policy can’t really do anything in the AD/NGDP area. So what do people like Larry Summers mean when they talk about a preference for using fiscal policy? … They are not recommending that fiscal policy be used to attack unemployment, they are recommending that fiscal policy be used to attack the private sector.
Krugman in his tribute to Milton Friedman (cough) explained differences in attitudes towards fiscal and monetary policy thusly…
He said it right there: fiscalists (he, DeLong, etc.) push fiscal policy not because of their economics but because of their politics — it is a rationale for getting more of the economy into the government’s hands, as they desire to achieve their political objectives.
(This seems also maybe a plausible enough explanation of why they deemed a fiscal policy of rising debt terrifying as “a looming threat to the federal government’s solvency” when it resulted from tax cuts, yet now take it as so benign, when resulting from spending increases outpacing tax increases, that Rogoff et al deserve ridicule for imagining that massive debt ever hurt any nation anywhere in their entire world-history data base. People tell me “You gotta give Krugman credit for admitting he was wrong about that”. Tell me any other time when Krugman has been happy to admit being wrong 🙂 )
14. January 2014 at 21:10
I love this post. I’ve been saying for a few years that the knowledge/calculation criticism against feasibility of efficient socialism is also the best argument against crude Keynesianism and so-called “fiscal policy”. There’s absolutely nothing about a low interest rate environment that should make centrally planned decisions driven by politicians more likely to create value than the decisions of decentralized individuals with actual skin in the game. None. And if a depressed economy makes it harder for entrepreneurs, what part of that makes it easier or more likely that politicians can do better or have better incentives? Nothing.
But then… crude Keynesians don’t seem particularly concerned with whether activity creates or destroys value in the first place. After all, Summers thought that the Kobe earthquake had produced “economic strength”. So they’ve got the entire point of economic activity upside down from top to bottom.
14. January 2014 at 21:11
“Fiscal policy can’t really do anything in the AD/NGDP area.”
This is a mischievous post by you, Scott, you know that Summers himself doesn’t believe that, socialism doesn’t have to be his primary motive.
I’m waiting for Morgan Warstler to comment on this, though, something tells me we’re about to see some exuberance…
14. January 2014 at 21:21
Supporting evidence on interest rates: http://www.cnbc.com/id/101332291
And what do you think about this discussion?
http://marginalrevolution.com/marginalrevolution/2014/01/which-countries-will-have-the-next-financial-crisis.html
14. January 2014 at 21:26
OK, you may be right about what motivates some of the people arguing for fiscal stimulus. Sometimes people advocate positions for reasons other than the merits of the arguments for the position. Nevertheless, these people often manage to craft good arguments for their position, if only to convince others to support things they support for other reasons.
There are plenty of people with less interesting things to say than you who are perfectly capable of questioning peoples’ motives (and they do–some make a living out of it). I’d argue your comparative advantage, and the comparative advantage of other knowledgeable intellectuals, is to focus on making good arguments for your position, and criticizing arguments for contrary positions.
When I read things like “[s]ay you want fiscal policy because you think the economy needs more socialism and less capitalism”, I generally stop reading. You certainly engage in less of this kind of thing than many of your peers, and I commend you for that, but that doesn’t make this a good post.
14. January 2014 at 21:45
Ram, I’m not questioning anyone’s motives. Summers says that he favors bigger government even if monetary policy can keep AD on track. He’s very clear on that point. I believe his motives are exactly what he says his motives are. He’s honest.
As far as the fiscal vs. monetary issue, I was just trying to clarify things. There is no such thing as not using monetary policy to control AD. The only question is good and bad monetary policy, and whether you also want to do something else, like increase government spending. Even if you spend a zillion on infrastructure, you still need the monetary instrument set at a level expected to produce on target AD.
Saturos, Financial crises are unpredictable.
14. January 2014 at 21:49
Extended comments from Summers: http://www.washingtonpost.com/blogs/wonkblog/wp/2014/01/14/larry-summers-on-why-the-economy-is-broken-and-how-to-fix-it
14. January 2014 at 21:55
“Everyone, If my monetary policy is a 5% NGDP growth target, then anything the government does to create 5% expected NGDP growth effectively means they’ve implemented my monetary policy.”
“MM is not about interest rates or the monetary base, it’s about expected NGDP growth. Get that on target and you’ve got a MM monetary policy.”
“If it causes the Fed to “see the light,” that’s great. But in the end of the day it’s up to the Fed.”
I think these three points are very important to your MM stance. Ultimately once the NGDP growth rate path is establish, it rests on the Fed to avoid tightening when the inflation component of NGDP starts to rise. This is the flaw in targeting stable inflation, by not allowing inflation to rise above their liking even if it’s associated with a steady NGDP growth path.
The question then is how should policy whether fiscal or monetary be used to establish a steadily rising NGDP growth rate and what are the distributional effects of the policy?
If the preferred method is by having the Fed increase the monetary base by acquiring treasury securities, at some point in time those treasury securities came into existence when the treasury issued them as a result of a government budget deficit. Even if the answer is to target NGDP by having the Fed conduct QE/OMO, the question as to how the budget deficit should manifest itself would need to be addressed.
“You don’t favor fiscal stimulus; you favor more government spending. And not transfers, those are tax cuts too. You favor more government OUTPUT.”
I think you make some excellent points at times, but this statement is far too divisive. Someone that suggests that the budget deficit should be larger does not necessarily support having government actually produce real output. Furthermore it becomes problematic when individuals use words different. I personally hate the word “stimulus” in general whether applied to fiscal or monetary policy.
As from calling transfers tax cuts, I see your reasoning for this, but some don’t associate the word transfers with tax cuts but with spending. I think of them as spending, and as a result I certainly don’t associate all spending especially direct transfers to individuals as favoring more government output.
So this leads me to a few thoughts.
1. The MM stance is to establish and maintain a 5% NGDP growth path. This would effectively replace the Fed’s inflation target.
2. This most likely would or may involve the Fed conducting OMO/QE to get us to that path which involves increasing the monetary base by acquiring treasury securities.
3. Treasury securities would/may need to be issued which involves running a budget deficit.
4. The question than is how should the budget deficit come into existence? Imagine a scenario where the Treasury issues securities and receives deposits (the Fed credits their reserve account and reduces the banking system’s balance). The Fed then acquires the newly issued treasury securities in the secondary market. How should the government spend their resulting balance? Would issuing checks to every citizen in equal amounts be favoring government output? Would this be pro-private/household sector? Doesn’t this place the spending decision in the hands of the citizens receiving those checks instead of the government bureaucracy? I can see why this can be like a tax cut, but tax cuts are as evenly distributed as they favor those in the formal workforce.
Ultimately, don’t we all just want to improve upon the situation we find our society in? I think it is easy to lose sight of that when discussing how policy should be used. Ultimately, I’m in favor of whatever works while making sure the distributional affects which come from both monetary and fiscal policy are as fair as possible.
15. January 2014 at 00:37
“He says the private sector will spend money on wasteful investments when rates are low, so it would be better for the government to spend the money.”….on wasteful investments.
15. January 2014 at 03:04
Scott,
You post contains a number of assertions that I find bizarre. Perhaps you can elucidate in future posts. E.g…
Your first paragraph claims that monetary and fiscal policies are alternatives only in barter economies. I’m completely lost.
Your second paragraph claims that monetary and fiscal policy are the same thing. Well my definition of those two policies ties up with the Oxford Dictionary of Economics definitions, and no way are they same thing.
Your third paragraph starts “Fiscal policy can’t really do anything in the AD/NGDP area.” Well as you know, there’s plenty of debate about that. The consensus is that while fiscal policy is thwarted to some extent by crowding out, it’s not totally thwarted. But a better solution to that problem, I think, is to fund fiscal stimulus via printed money rather than borrowed money (as Keynes suggested). I.e. go for a mix of monetary and fiscal policies.
Re your claim that Keynsians are left of centre, maybe they are. But that doesn’t prove that Keynsianism as such is left of centre. Keynsianism (or at least the form of Keynsianism I support) essentially just advocates deficits in a recession: it says nothing about whether that comes in the form of more government or more household spending.
It may be that chemists tend to be left of centre, but that doesn’t prove that chemistry is inherently left of centre.
15. January 2014 at 03:15
Scott,
Quick question. Suppose the Fed policy goal was for the base to grow at a fixed and constant rate, and they didn’t give a hoot about output or prices. In this scenario, could increased government spending and/or tax cuts cause an increase in AD?
15. January 2014 at 03:32
Scott,
is it fair to say, that at some point you also prefer socialism to markets?
I remember that you said in case that bonds and money became equivalent (negative interest rates forever) and central banks were not aloud to buy private assets, you would advocate state funds (the government investing in the stock market) financed through monetary policy, accusing my preferred solution (giving money to individuals through helicopter drops) as wastefull. Isn’t this exactly Summers logic? Government knows best?
I also remember that you said helicopter drops are welfare while advocating wage subsidies. And you seem to be in favor of government forcing people to save (Singapore style retirement saving accounts)…
Is it fair to say that (from my libertarian perspective) there is a heavy dose of paternalism in your thinking? In contrast to Milton Friedman by the way….
15. January 2014 at 04:23
Scott – “Fiscal policy can’t really do anything in the AD/NGDP area”
Wow. Do you mean just that monetary policy can offset a fiscal boost, or that fiscal policy can’t achieve a boost in the first place?
15. January 2014 at 05:20
Anders,
I’d put it a third way: fiscal policy cannot accomplish anything in that area that cannot be accomplished via monetary policy.
15. January 2014 at 05:21
(It’s analogous to reserve requirements vs. OMOs in monetary policy: central banks could use reserve requirement changes rather than OMOs, but without any improvement and some technical problems.)
15. January 2014 at 05:22
(Triple post mania: the above should be read as referring to normal circumstances.)
15. January 2014 at 05:30
Here is a recent tweet by Andy Harless:
“Another possible definition of sec stag: r*<<g, so equilibrium asset values are over-sensitive to subjective risk premia"
My impression was that Summers was at least partially using this argument too. Basically they argue that with fiscal stimulus we would get markets that are more stable, and accordingly AD would be more stable too.
I am not persuaded by this argument yet, as it ignores fiscal tail risks.
15. January 2014 at 06:15
Mike, if Scott’s right, then we could have gotten out of the recession years ago if Obama, Krugman, etc. had clearly followed the following strategy.
1) We need to try unconventional monetary stimulus to get the economy moving. If not offset by the fed, an employer side tax cut would also be a good idea.
2) As a related issue, while long term rates are low, this is a good time to finance the following positive value government projects: a, b, c.
Then even if (2) was politically infeasable, at least you could exert pressure in the direction of (1). Now it’s true that Obama and Krugman want a bigger governement than I do, but presumably we all wanted the recession to end.
If Scott’s not right, then (1) would be a waste of time, but if he is, then Obama left money on the table by giving up on monetary policy early on, and the best case you can make for Krugman is that he failed to clearly articulate that unconventional monetary policy might work and was worth trying.
15. January 2014 at 06:33
Rafael, You said;
“This most likely would or may involve the Fed conducting OMO/QE to get us to that path which involves increasing the monetary base by acquiring treasury securities.”
No, NGDPLT would not require a larger base. And I do not believe that a sound monetary policy has important distributional effects.
You are right that I should have been more careful about describing the proponents of fiscal policy, I was thinking specifically of those taking the line that Larry Summers takes.
libertaer, You said;
“Isn’t this exactly Summers logic? Government knows best?”
No, they are very different. I favor having the Fed buy stocks if the only alternative is depression. (Of course it’s not the only alternative, but I meant that it was my preference if they weren’t willing to have a higher NGDP target path.) Summers favors public spending over private spending even when the economy is at a full employment. Helicopter drops require higher MTRs, which is worse than government owning index funds for short periods.
Anders, It can’t do anything if the central bank is targeting a nominal aggregate. A monetary policy where future expected AD is expected to be on target is a necessary and sufficient condition for success.
Vaidas, Is it true that more stable asset prices lead to more stable NGDP? I doubt it. The stock market crash of 1987 had no impact on NGDP, and it was as large as the 1929 crash. I think what throws people off is that NGDP shocks often cause asset price instability. But when the asset movement is exogenous (1987) it has almost no effect.
15. January 2014 at 06:41
@W. Peden
You say any boost that fiscal policy can do, monetary policy can also do. But fiscal policy adds to people’s net worth; monetary policy can’t do that.
A cut in the central bank rate decreases the amount of interest income which the govt is paying to the non-govt sector. This worsens the non-govt sector’s balance sheet relative to the counterfactual.
I acknowledge that a cut in rates may boost the disposal income of a net debtor household with floating rate debt. But it will also lower the disposable income of a net creditor household with floating rate assets (eg pensioners). I’ve never understood why those who favour monetary policy think it’s appropriate to boost aggregate demand (or NGDP, if you prefer) by arbitrarily redistributing income between different cohorts of the population.
15. January 2014 at 06:50
Scott,
Great post.
Jim Glass,
Great find.
15. January 2014 at 07:13
And of course there is no empirical evidence that fiscal policy has ever boosted aggregate demand. In every instance you might point to where fiscal policy was “expansionary” and “worked”, it was accompanied by expansionary monetary policy. The reverse is not true. Scott’s favorite example, the 1937 devaluation, is only the most dramatic U.S. example of this. The best current example is Japan.
I noted in a comment a couple of days ago some of the same things Scott says here. In particular, note that advocates of fiscal policy almost never recommend tax cuts, even though the theory they claim to believe in says tax cuts work about as well as higher government spending does, and practical considerations (the lack of “shovel-ready” projects) also argue for tax cuts rather than increased government spending.
What Scott is saying here should not be controversial. It was conventional wisdom in the economics profession before 2008, and it’s only because the best-known economists are political liberals that there is any public debate.
15. January 2014 at 07:36
Daniel Kuehn:
“I find Sumner so baffling sometimes. A lot of his criticisms of non-market monetarists are based on strange diagnoses of their claims. Until this morning reading this post I would have thought it obvious that most Keynesians see Japan in the 90s and the Bernanke Fed as running into exactly the same problems. And yet Sumner’s critique here is largely that Keynesians view them differently (and that’s a problem). Sumner claims that in the 1930s they would have laughed at Friedman and Schwartz, but monetary tightening was a major part of the diagnosis of the Depression for Keynes. Where I think the real disagreement lies is in the claim that the Fed could have saved us. Keynesians then and now think that is a harder case to make – not that monetary policy is useless or impossible (another thing Keynesians don’t claim but Sumner has a nasty habit of suggesting they do), but that it’s a difficult tool to use in deep depressions. Once a monetary collapse happens, long-term expectations can be hard to dislodge (market monetarists tend to think they’re easy to change, and therefore if they are not changed the monetary authority must be engaging in tight policy).”
http://factsandotherstubbornthings.blogspot.com/2014/01/sumner-on-far-sighted-keynesians.html
15. January 2014 at 07:58
@ Jeff “there is no empirical evidence that fiscal policy has ever boosted aggregate demand”
And there is empirical evidence that monetary policy has boosted aggregate demand?
Of course, in reality there cannot be empirical evidence for either fiscal or monetary policy since we never know the counterfactual of a given policy action. The vast majority of recent cyclical peaks and troughs have seen both fiscal and monetary policy.
Logic, rather than evidence, tells us that a monetary ‘stimulus’ will lower the private sector’s net worth.
As Ralph said above, don’t confuse the best-known advocates of fiscal policy with fiscal policy itself. The archetypical advocate of MMT (a post-Keynesian school) is Warren Mosler, a hedge fund manager, who advocates tax cuts more than spending increases.
15. January 2014 at 08:03
Based on your EconLog “bubbles” piece you linked in the previous post, it seems you agree that there are market inefficiencies (or failures of regulation) that could cause low real rates to boost asset prices without creating a corresponding boost to investment. That seems to be Summers’s point. Or am I misreading?
15. January 2014 at 08:28
“Then I’d say you’ve just adopted my preferred monetary policy””you set monetary policy at a position where AD was expected to grow at the socially desirable rate.”
You seem to be claiming that a successful fiscal policy is just monetary policy in disguise. To the extent that slow NGDP growth is caused by increased demand for money then this may be true.
But what if slow NGDP growth is caused by a desire by people not just to keep more of their wealth in cash but by a desire to increase savings as a hedge against future loss of income ? In this case fiscal policy addresses the issue by increasing people’s income directly and allowing them to maintain spending at the desired level as well as increasing savings.
Monetary policy may not be able to do this as easily since as it swaps assets for money the price of assets will increase in money terms , but the expected returns on those assets will fall (with lower interest rates). This will impact the effect of monetary policy on AD since people base their current spending on expectations of future income stream and not just on wealth.
If you use fiscal policy to adjust nominal income and wealth levels to that required for health NGDP growth, you can then leave monetary policy to do what its good at: maintaining the correct balance between money and other assets.
15. January 2014 at 08:58
“No, NGDPLT would not require a larger base. And I do not believe that a sound monetary policy has important distributional effects.”
So how would you advise the Fed and/or Treasury to hit a NGDPLT especially in a situation where that target is currently not being met? If via expectations, what happens when expectations start to diverge from measured outcomes? Although clearly not NGDPLT, the Fed has a 2/2.5% inflation target. The Fed has missed this target and expectations are that inflation will be below this target going forward. What can the Fed do to not only change expectations, but make sure that the desired result actually occurs?
As you mention changing the monetary base is not required.
“Helicopter drops require higher MTRs, which is worse than government owning index funds for short periods.”
Why do helicopter drops (direct transfers to individuals/households) require higher MTRs (marginal tax rates)? The transfers would result in a budget deficit which could be financed issuing t-securities. Equal transfers to all individuals are like tax cuts, but tax cuts that have less of a distributional effect since it goes to those not only formally employed, but also to those who are not.
Anders:
“But fiscal policy adds to people’s net worth; monetary policy can’t do that.”
If monetary policy results in a portfolio rebalancing effect or change in market psychology where equity prices rise, does this not at least temporarily add to people’s net worth? If yes, then monetary can add to people’s net worth. However wouldn’t this policy distributionally favor people holding assets which are rising in price relate to others (those that don’t or mainly depend on income from labor)?
Jeff:
“In particular, note that advocates of fiscal policy almost never recommend tax cuts, even though the theory they claim to believe in says tax cuts work about as well as higher government spending does”
I’m not sure which advocates of fiscal policy you are referring to, but I concur with Anders. I’ve read too many views on how to use the government budget to draw the conclusion that some advocates of fiscal policy don’t recommend tax cuts. Personally, I favor equal direct transfers to all (financed by issuing t-securities not raising MTRs) since tax cuts favor those receiving the cuts relative to those who aren’t.
15. January 2014 at 09:08
More often than not, what they’re really advocating is increased non-market resource allocation. But it’s debatable whether fiscal advocates view this as a side-benefit, or one of the main benefits.
15. January 2014 at 09:14
Alex Alters,
Is that always true? Advocates of running a budget deficit/treasury issuance to facilitate direct transfers to individuals place the resource allocation decisions (spending decisions) on the receivers of the transfers. The receivers would go through the market. Essentially, the government is transferring the spending decision which the government bureaucracy would have to otherwise make.
15. January 2014 at 11:10
Scott,
One more time. Just trying to clarify your views. Suppose the Fed policy goal was for the base to grow at a fixed and constant rate, and they didn’t give a hoot about output or prices. In this scenario, would increased government spending and/or tax cuts cause an increase in AD?
What I’m trying to get at is if you believe that absent a monetary policy offset, whether increased government spending and/or tax cuts will boost AD.
15. January 2014 at 11:43
Why not monetize the US debt? It seems like a win-win. Replace the money supply lost because of the panic to improve job growth and wipe out trillions of debt. Why not print! print! print! until NGDP is consistently growing by 6% ? A generation ago, a low 3% inflation was beyond our dreams. Now it seems a that inflation as high as 3% is beyond our dreams. Eliminate the debt and all that interest money will be available to double infrastructure.
15. January 2014 at 11:45
Daniel Kuehn:
“Sumner claims that in the 1930s they would have laughed at Friedman and Schwartz, but monetary tightening was a major part of the diagnosis of the Depression for Keynes.”
Actually Keynes was very slow to acknowledge that monetary tightening may have been a problem, and steadfastly refused to advocate that the UK devalue or detach from the gold standard.
Anticipating the Great Depression? Gustav Cassel’s Analysis of the Interwar Gold Standard
Douglas A. Irwin
November 2011
Pages 28-32 are most relevant. Here are some highlights:
http://www.dartmouth.edu/~dirwin/Cassel.pdf
“…In a January 1929 article on the League of Nations gold inquiry, Keynes found himself coming around to Cassel’s view of the international gold problem. Keynes (1981 [1929], 776) wrote that “Professor Cassel has been foremost in predicting a scarcity” of gold, adding that “I confess that for my own part I did not, until recently, rate this risk very high.”…”
“…Although this article demonstrates Keynes’s familiarity with Cassel’s arguments, in his subsequent writings Keynes hardly mentioned the international gold problem at all. Keynes rightly believed that the international monetary cooperation that Cassel had always demanded was simply not going to happen, so he looked for other solutions. However, these solutions did not including leaving the gold standard. Despite the fact that he had been an opponent of the gold standard in the early 1920s – calling it a “barbarous relic” in his Tract on Monetary Reform (1923) – and although he welcomed Britain’s departure from the gold standard when in finally occurred, Keynes steadfastly refused to advocate a British devaluation or departure from the gold standard.19 In his testimony before the Macmillan Committee in 1930, Keynes concluded that the costs of departing from the gold standard outweighed the benefits because the burden of servicing Britain’s short- term foreign currency debts would increase by the amount by which the pound fell in value. Furthermore, he argued, abandoning the gold standard would be a breach of faith with Britain’s creditors, a violation of trust that would damage London’s reputation as a financial center…”
“…However, Keynes began to change his view when monetary policy was no longer handicapped by golden fetters: after Britain left the gold standard in September 1931, interest rates came down but unemployment remained high. According to Patinkin (1982), this is when Keynes became a monetary policy skeptic and began to push for increased government spending on investment as a way to get the economy moving again. As Keynes stated in late 1931, after Britain left gold, “I am not confident . . . that on this occasion the cheap money phase will be sufficient by itself to bring about an adequate recovery of new investment. . . . . If this proves to be so, there will be no means of escape from prolonged and perhaps interminable depression except by direct State intervention to promote and subsidize new investment” (Keynes 1982, 60)…”
“…Because Keynes judged the stance of monetary policy largely if not exclusively by interest rates, with low rates indicating to him monetary ease, he became skeptical of the value of monetary policy as a stabilization tool when nominal rates were so low.
By contrast, Cassel never lost faith in the power of monetary policy to improve economic conditions…”
To me it seems like deja vu all over again. Today Krugman is similarly influential, and is supportive of unconventional monetary policy after the fact, but never really advocates it, because his belief that monetary policy mainly works through the traditional interest rate channel makes him skeptical of unconventional monetary policy’s benefits in a liquidity trap.
15. January 2014 at 12:01
Mark,
Very interesting post. I tend to think PK’s reasons are more political. And BTW – I think we have had a ZLB-like problem, because the Fed has artificially allowed bank reserves to soar. There is no transmission mechanism that will work if Fed action simply results in higher reserves.
15. January 2014 at 12:25
Don
“Why not monetize the US debt? It seems like a win-win. Replace the money supply lost because of the panic to improve job growth and wipe out trillions of debt. Why not print! print! print! until NGDP is consistently growing by 6% ?”
Treasury securities (the US debt) held by the private sector are the assets of those holding them. Why would having the Fed acquire them in exchange for reserves (the non-bank seller would receive bank deposits) directly lead to higher inflation? Isn’t inflation a result of people with deposits using them to acquire goods and services which causes the sellers to increase prices (if they can)? Why would aggregate prices immediately change first and not output? Secondly, what if the deposit holder decides not to spend the new deposits that way?
“Eliminate the debt and all that interest money will be available to double infrastructure.”
The interest paid on treasuries results in a gain to the private holder of those treasuries. Why would reducing their interest income allow the government to increase their spending on infrastructure? I am not saying that they should one way or another, but they could issue new treasuries now at low rates if they wanted to spend on infrastructure. To not be doing so is a political choice.
15. January 2014 at 13:16
o nate, Yes, but unlike Summers I don’t view asset prices “bubbles” like 1987 as a problem.
The Market fiscalists, You said;
“If you use fiscal policy to adjust nominal income and wealth levels to that required for health NGDP growth, you can then leave monetary policy to do what its good at: maintaining the correct balance between money and other assets.”
No, monetary policy is good for keeping NGDP growing at a steady rate—nothing more, nothing less.
Rafael, You asked;
“So how would you advise the Fed and/or Treasury to hit a NGDPLT especially in a situation where that target is currently not being met?”
I would tell them to adjust the monetary base until expected NGDP growth equaled the target. That would probably require a reduction in the base. Your comment about inflation targeting is not relevant, as level targeting is radically different from inflation targeting.
You asked:
Why do helicopter drops (direct transfers to individuals/households) require higher MTRs (marginal tax rates)?”
You need to raise taxes in order to have the funds to buy back all that base money once the demand for money normalizes. Otherwise you get hyperinflation.
dtoh, You said;
“One more time. Just trying to clarify your views. Suppose the Fed policy goal was for the base to grow at a fixed and constant rate, and they didn’t give a hoot about output or prices. In this scenario, would increased government spending and/or tax cuts cause an increase in AD?”
Yes, it may well do so. But I don’t view a stable monetary base as “not using monetary policy.” It is a monetary policy. Nor do Keynesians view a stable monetary base as a neutral monetary policy.
Don, See my second reply to Rafael.
Mark, Excellent comparison between Krugman and Keynes.
15. January 2014 at 14:17
“You need to raise taxes in order to have the funds to buy back all that base money once the demand for money normalizes. Otherwise you get hyperinflation.”
So how do you see this unfolding?
1. Treasury issues t-securities by running a deficit and sends a check to all individuals.
2. The Fed acquires the treasuries by creating reserves.
3. The recipients use the checks to spend on goods and services increasing output and eventually aggregate prices.
4. This would suggest that the demand for money has normalized (demand for goods/services increase as demand for money falls). The helicopter drops/transfers stop or slow. 5. If this trend becomes unstable, the Fed can swap the t-securities on its balance sheet for the monetary base. The government could (should) raise taxes once demand has normalized or become excessive.
6. Tax receipts have almost definitely increased by this point due to the increase in economic activity/transactions for goods and services. The treasury uses the tax receipts to retire the treasury bonds held by the private sector.
Is this plausible? When does hyperinflation start to occur? What if the helicopter drops start very small and are progressively increased then decreased/reversed?
15. January 2014 at 15:26
I want fiscal policy. And I want tax cuts.
Let’s have local and state governments PERMANENTLY CUT sales/excise tax rates by 30%, across the board.
I guarantee you that this will stimulate the economy.
15. January 2014 at 15:37
Scott,
Don’t disagree at all. Having no central bank and no government controlled fiat currency would still be a deliberate monetary “policy.”
One last question and then I’ll stop bugging you on this issue. If you had a monetary policy which was “Do OMOs at a constant rate so that the base grows at 5% a year,” do you think it would be possible to hit a 5% NGDPLT target just by adjusting taxes and government spending (assuming a competent and nimble Congress)?
15. January 2014 at 16:00
“No, monetary policy is good for keeping NGDP growing at a steady rate””nothing more, nothing less.”
Here is a scenario where I don’t think it can do that without dangerously inflating asset prices:
– People fear a drop in future income so save out of current income to boost future income streams
– The CB expands base money but hits the zero bound before NGDP is back on target
– They continue to expand base money via purchases of assets that do have interest rates > 0.
– The qty of base money, asset prices and total financial wealth all rise but people don’t increase spending much because the lower interest rates they are getting mean that future expected income has not in fact risen. Eventually asset prices will be so high compared to final good prices that people will spend , but dangerous imbalances may have been created by then.
Fiscal policy avoids this problem by simply creating new money and new financial wealth directly and allowing monetary policy to focus on maintaining interest rate at a healthy level.
15. January 2014 at 16:40
To add to dtho’s inquiry if I may
“Do OMOs at a constant rate so that the base grows at 5% a year,”
If this is a strategy, would it be possible for the Fed to eventually run out of t-securities to acquire if say Congress decided to stop running deficits thereby not issue any new treasuries? What if Congress started running large annual surplus with the goal of reducing the quantity of treasury securities outstanding (aka paying off government debt)? How would the Fed need to adjust?
15. January 2014 at 17:11
” And don’t call it fiscal stimulus either. Tax cuts are fiscal stimulus, and as you might have noticed almost all the Keynesians favored the tax increases Obama adopted a year ago. You don’t favor fiscal stimulus; you favor more government spending. And not transfers, those are tax cuts too. You favor more government OUTPUT. You’ve rejected neoliberalism, so say so.”
If your demands are that easy I’m just going to have to go ahead and oblige them. I’ve rejected neoliberalism. What real Keynesian should be ashamed to admit that?
This hypothetical socialist Monetarist hasn’t rejected it I’m guessing?
Also it’s not true that Keynesians are always for tax hikes just for the rich. Last year Keynesians supported the end of the Bush tax cuts for the rich. However, most wanted the payroll tax holiday to continue.
Of course, God’s party-and yours-didn’t want that. It’s not that Keynesians hate tax cuts and conservatives love them but that both like certain types of tax cuts and certain types of tax hikes.
Conservatives-or Neoliberals if you prefer-like to cut taxes for the rich and raise them for the poor-the consumption tax, value-added tax, et. al.
15. January 2014 at 18:40
Rafael,
It doesn’t matter what assets the Fed buys. Treasuries are convenient… that’s all. They could just as well buy bank debt, corporate bonds, auto loans, or credit card receivables.
15. January 2014 at 20:50
dtoh,
I believe that is illegal.
http://www.law.cornell.edu/uscode/text/12/355
“Every Federal Reserve bank shall have power…
To buy and sell in the open market, under the direction and regulations of the Federal Open Market Committee, any obligation which is a direct obligation of, or fully guaranteed as to principal and interest by, any agency of the United States.”
Also: http://www.federalreserve.gov/faqs/what-are-the-federal-reserves-large-scale-asset-purchases.htm
Secondly, by acquiring privately issued securities the Fed would be favoring that issuer. This would result in direct distributional outcomes that do not occur when the Fed purchases t-securities. Furthermore, by acquiring such securities this would expose the Fed’s balance sheet to credit risk. A potential loss would result in the Fed remitting less interest to the Treasury which would increase the budget deficit. Furthermore, if the privately issued security is credible the interest paid to the Fed would result in a larger remittance to the Treasury (since it would offer a higher rate than t-securities) which would be a relative loss of income to the private sector.
If the above is accurate, I’d have to respectfully disagree with you dtho. It certainly does matter.
15. January 2014 at 23:29
Rafael,
So change the regs. Other CBs buy private assets.
Risk adjusted return on private assets will be same as Treasuries. Over the long run losses will match higher interest earned. Short term it doesn’t matter. Any losses would be a small fraction of the government budget. On a risk adjusted basis there would be no relative loss of income to the private sector.
Fed doesn’t need to favor an issuer. They could conduct an auction or buy a basket of assets.
These are all trivial issues.
16. January 2014 at 03:21
Scott:
“Is it true that more stable asset prices lead to more stable NGDP? ”
Theoretically yes. Monetary policy has to be more active to stabilize NGDP when asset prices are not stable, and this is costly.
“The stock market crash of 1987 had no impact on NGDP.”
Greenspan has eased the policy after the crash. The question is – were NGDP expectations stable during the crash of 1987? Probably not.
16. January 2014 at 04:00
Scott, Rafael,
“You need to raise taxes in order to have the funds to buy back all that base money once the demand for money normalizes. Otherwise you get hyperinflation.”
Why can’t the CB just pay interest on reserves? You don’t need assets (treasuries or private assets) for vacuum cleaner operations, just raise the interest rate on base money.
You “print” money to suck up “printed” money, that’s genius! 🙂
16. January 2014 at 06:21
Rafael, If you triple the monetary base and IOR is zero (your assumption) then the hyperinflation begins when interest rates rise above zero, or perhaps even when they are expected to rise above zero.
dtoh, You might or might not be able to hit a 5% NGDP target, it depends on the exact situation.
Market fiscalist, There are a lot of assumptions in that chain of reasoning that are highly debatable. The odds of them all being true is remote.
Rafael, What sort of assets did the Fed buy during the crisis of 2008?
Mike, I knew you weren’t a neolib, I was thinking of people like DeLong.
Vaidas, That may be, but 1987 suggests to me that any effects in that direction are of trivial importance.
Libertaer, You need to raise taxes to get the revenue to pay IOR.
16. January 2014 at 06:57
“Fiscal policy avoids this problem by simply creating new money and new financial wealth directly”
If this is true we should all be Communists by now. But we aren’t, because in truth the government tends to horribly misallocate resources.
16. January 2014 at 07:00
If the monetary base has tripled, how would the Fed’s target interest rate rise above zero without the Fed selling securities from its balance sheet in the first place which would drain reserves?
Furthermore, in the steps I provided I include two ways that a reserve drain would occur. They would happen after the demand for money has declined (demand for goods/services rises), which would cause output and eventually prices to rise. The reserve drains were outlined in the steps. Wouldn’t you need to factor in the rate of GDP growth (the taxes that derive from that economic activity) into your analysis involving reserves and treasuries held by the private sector?
Also, in the scenario the monetary base was never mentioned to have tripled. The Fed wouldn’t have to buy the treasury securities unless they felt it helped them hit their goals when using monetary policy. If the monetary base grows to the point where it is three times the size it is today, wouldn’t that signify that the economy is still probably pretty weak and interest rates are still depressed?
Considering the Fed are preparing to use a reverse repo facility to strengthen the floor on overnight interest rates, we aren’t even close to the moment where overnight rates rise materially above zero. The economy can’t just move from this condition to hyperinflation without the Fed noticing right? So as the monetary base is expanding, I’d like to think the Fed would have the ability to maneuver their balance sheet to avoid any hyperinflation scenario.
16. January 2014 at 07:32
“Fed doesn’t need to favor an issuer. They could conduct an auction or buy a basket of assets.”
In either of those cases it would be impossible for the Fed to be all inclusive. If the Fed acquires a basket of assets, those holding the securities or issuing securities that are in that basket are made better off relative to those holding/issuing securities that don’t make it into that basket. In the Just Banking conference of 2012, Adam Posen and Richard Werner both discuss the possibility of having the Fed buy different types of securities. If I remember correctly Posen was against this for some of the reasons I’ve mentioned. Werner wanted to set up new facilities which would issuer securities that the Fed would acquire. The facility would then lend to small business. I believe Posen did not want the Fed in the business of making such decisions which would result in distributional outcomes.
“Rafael, What sort of assets did the Fed buy during the crisis of 2008?”
Surely certain regulations have to be made flexible in a crisis period in order to allow the Fed to save the system. If it is a choice between buying private securities in a time of crisis or depression, let’s go with buying private securities! Katharina Pistor’s paper discusses this.
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2178066
“So change the regs. Other CBs buy private assets.”
At that point, why not just change the laws so that the Fed can directly instruct banks to credit all private bank accounts equally? The Fed would credit the bank’s reserve balances. The Fed’s equity would take a hit, but this would result in a larger budget deficit which the treasury would cover by issuing T-securities. At least this way the Fed can directly impact all citizens more equally rather than having to select which securities go into the basket. I’d imagine this would be a much fairer monetary policy considering how concentrated asset holders are. Why favor any asset holders at all? At least when the Fed buys t-securities, it is up to congress to use their budget fairly on all our behalf.
16. January 2014 at 08:18
Rafael,
“If the Fed acquires a basket of assets, those holding the securities or issuing securities that are in that basket are made better off relative to those holding/issuing securities that don’t make it into that basket.”
In general the price of all assets including those not in the basket will rise equally. You should think of all financial assets as being fungible.
As I have discussed elsewhere, helicopter drops if done regularly will just cause a change in the price level with no change in real output. And if this were not case, what would stop people from hoarding cash or money in order to become the beneficiary of future drops. (If on the other hand what you mean by helicopter drops is just a tax credit or negative tax for everyone, then that’s fiscal policy not monetary policy, and IMHO it is politically unlikely to happen.)
It’s highly unlikely in our lifetime that the Fed is going to run out of TSYs to buy especially if they take steps to reduce excess reserves held by banks.
16. January 2014 at 08:19
Scott,
“You might or might not be able to hit a 5% NGDP target, it depends on the exact situation.”
What would it depend on?
16. January 2014 at 08:48
dtoh
“As I have discussed elsewhere, helicopter drops if done regularly will just cause a change in the price level with no change in real output.”
Why wouldn’t firms respond to increased demand generated by heli-drops by increasing output before raising prices especially in the situation we are where we have unused productive capacity? If firms were able to respond by raising prices instead of increasing output, wouldn’t this be a sign of an uncompetitive market environment? We would see business profits rise as prices rise. Wouldn’t this incentivize competition for those profits by either enticing firms to enter and undercut incumbents by offering offering lower prices (thereby putting downward pressure on the price level)?
“And if this were not case, what would stop people from hoarding cash or money in order to become the beneficiary of future drops.”
Nothing. However, what is the point of a hoarding deposits for the sake of hoarding deposits forever? Isn’t the goal of hoarding (accumulating purchasing power)to eventually spend them? Also, shouldn’t we factor in the varying marginal propensities to consume? I doubt that 70-90% of the population who are money constrained would just sit on those deposits? If they do, wouldn’t this improve their financial position especially after the financial crisis? Wouldn’t this reduce their dependence on credit?
“In general the price of all assets including those not in the basket will rise equally. You should think of all financial assets as being fungible.”
So how does this directly benefit the majority of individuals who hold a negligible amount of financial assets relative to the top 10% of the population? Wouldn’t this benefit those holding financial assets vs those who don’t (or own little)?
16. January 2014 at 09:08
Scott: “That may be, but 1987 suggests to me that any effects in that direction are of trivial importance.”
Ex post costs of 1987 stimulus were low, but one can argue that ex-ante they were high. Ex post costs may be high too if moral hazard costs are included.
However, I don’t see how fiscal stimulus helps with any of this, as treasuries will get riskier if we do more fiscal stimulus.
16. January 2014 at 09:53
Rafael,
I wrote some comments under the post “dtoh on DeLong.” Have a look at those.
“So how does this directly benefit the majority of individuals who hold a negligible amount of financial assets relative to the top 10% of the population? Wouldn’t this benefit those holding financial assets vs those who don’t (or own little)?”
Fed action has two effects on rates (asset prices). The actual purchase pushes up prices, but if it changes expectations of future NGDP, it raises the expected Wicksellian rate pushing real prices down. The net effect is indeterminate. (Most likely nominal bond prices will go down and nominal equity prices will go up).
If the Fed is doing it’s job, however, NGDP expectations will stay constant so you don’t get into this situation. If the Fed does not do it job, then gains earned by higher asset prices will have been offset by the initial losses incurred when the Fed let NGDP expectations drop.
Not to be presumptive, but I think monetary policy is hard to understand if you think about OMO as some how giving money away. It’s a lot easier to think about it as an exchange where something of equal value is tendered for the money.
16. January 2014 at 11:45
“Fed action has two effects on rates (asset prices). The actual purchase pushes up prices, but if it changes expectations of future NGDP, it raises the expected Wicksellian rate pushing real prices down. The net effect is indeterminate. (Most likely nominal bond prices will go down and nominal equity prices will go up).”
I agree with this. The net effect is indeterminate given that a rise in the natural rate of interest will also have an affect on asset prices. The actual purchases put downward pressure on rates, but a rise in the natural rate will put upward pressure on rates. Could this be what we are seeing since the middle of 2013 where nominal bond prices sold off while equity prices have risen (as you describe)?
“If the Fed is doing it’s job, however, NGDP expectations will stay constant so you don’t get into this situation. If the Fed does not do it job, then gains earned by higher asset prices will have been offset by the initial losses incurred when the Fed let NGDP expectations drop.”
What should the Fed actually do to make sure NGDP expectations remain constant? I agree with the idea that if NGDP expectations decline then the gains earned by higher asset prices will have proven to be temporary as prices start to trend downward. Why would NGDP expectations drop? I’d think expectations would change if/when actual measured NGDP comes in less than what was expected. If observed NGDP comes in less than what was expected, and the answer is to increase asset purchases, doesn’t it come back to what should the Fed purchase?
“but I think monetary policy is hard to understand if you think about OMO as some how giving money away. It’s a lot easier to think about it as an exchange where something of equal value is tendered for the money.”
It isn’t giving ‘money’ away. What is the actual mechanism of OMO/QE? The Fed increases its balance sheet by swapping reserves for government guaranteed securities (zero credit risk). If the Fed acquires the t-bonds directly from the banking system, the Fed credits the bank’s reserve balance in an equal amount to the securities sold. If the Fed acquires the securities from non-banks/dealers, the Fed still credits the banking systems reserve balance while the bank credits the sellers deposit balance within the banking system. As you describe this is an asset swap of equal value where the net worth does not directly change.
This is why which securities the Fed purchases are important because the securities the Fed acquires would have had to have been initially issued by some entity. The treasury issues securities on our collective behalf. Issuers of privately created securities create them on their behalf. If the Fed decided to purchase a specific set of securities (even a basket), the issuers of those securities would benefit as they would have the exclusive ability to issue new securities of that kind. It would be rational for securities dealers or investors to demand those newly issued favored securities as they would then be able to sell them on to the Fed as the risk of not find a buyer at the prevailing price would be lower.
Therefore, as long as the Fed is only able to deal in government guaranteed securities (which if the logic above is correct is the most neutral LSAP and would rely on Congress to decide on direct distributional outcomes via the budget process… I think this is the point Posen makes) when outside a crisis period, then the Fed’s ability to impact expectations of NGDP by adjusting the monetary base would be tied to the treasury securities held by the private sector. In the event where Congress decided to run balanced budgets or surpluses (not sure if this is even sustainable) the treasury would stop issuing or retire treasury securities held by the private sector which impacts the quantity available for the Fed to purchase. Thus, it is imaginable that at some point the Fed could potentially run out of securities it could legally purchase.
As for the post “dtoh on Delong” (have yet to read through comments will later):
I agree with you that monetary policy at the ZLB still matters (it always matters)! However so does treasury issuance (the residual of fiscal policy) as long as Fed is legally constrained in what they can buy when conducting LSAPs. The monetary off set is tied to the Fed’s ability to purchase treasuries. If the Fed can start acquiring privately created securities (outside of a crisis period) then I’d love for them to buy my securities (how can we get that to happen!?)!
17. January 2014 at 00:49
Rafael
All asset prices will go up (or down if the the NGDP expectation effect is larger) equivalently at the same time. Doesn’t really matter what the Fed buys. The only exception is if you have an illiquid issue or the Fed buys a lot of one particular issue, but that doesn’t impact the rate at which the issuer can issue new securities. It doesn’t really matter who owns the existing paper, the price of a new issues will depend primarily on the issuer’s ability to service the debt or generate returns on equity.
That said if the Fed bought a high percentage of the outstanding amount of a particular asset class it could change the price of that financial asset relative to other financial assets, but with a minimal amount of planning this would not happen…. and if you were really worried about it, you could just set up a lending facility to banks who would just intermediate the additional supply of credit and allocate it based on risk/price just like they always do.
Again, it makes it easier if you think of all assets as being completely fungible. It is also easier if you think of the Fed and banking system as a single entity. Someone (FOMC) pushes a button and it makes the price of credit (financial assets) go up or down for everyone.
BTW – If you have a mortgage, the Fed might already own your securities.
17. January 2014 at 03:50
“Libertaer, You need to raise taxes to get the revenue to pay IOR.”
Why? The CB can just print the money.
That’s one point I learned from Market Monetarism, monetary policy has nothing to do with banking, there are no assets or liabilities, no need for funding. Ideally we should stop thinking of the money supplier as a bank. An apple producer is not a bank either.
The money supplier could target NGDP by raising the money supply through helicopter drops (printing and giving away) and lowering it through vacuum cleaner operations (paying interest on base money). The interest payments are done in “freshly printed” base money, no need for taxes. So even when you want to contract the money supply, you are raising the base. The contraction is done by a fall in velocity. If NGDP=mv, then helicopter drops raise m leading to a rise in v, while interest on base money raises m a little too, but lowers v much more.
In general, you can raise and lower NGDP by rasing and lowering m and/or v. That gives you four options:
1. rasing m and v through heli drops
2. lowering m and v through taxes
3. rasing v and lowering m through negative interest rates (not possible with currency)
4. rasing m and lowering v through positive interst rates
Forget asset purchases! You are still thinking via the banking model. Monetary policy has nothing to do with banking.
17. January 2014 at 07:26
“Again, it makes it easier if you think of all assets as being completely fungible.”
Every security has to be held by someone. The price of that security would depend on the bid/ask in the marketplace. The bid/ask spread is a function of security’s dealers/market makers. When prices are falling, the security’s dealer may begin to accumulate securities as inventory to sell once demand for those securities begin to rise again. The dealer will fund these asset purchases with money market funding. Perry Mehrling gets into all of this in his writing/class.
The Fed played a critical role in the crisis as dealers began stepping away from certain security’s markets. They didn’t/couldn’t hold enough inventory and as a result market prices especially for certain securities were hard to determine. Which securities were highly in demand? T-securities. The fungibility of those assets would depend on market liquidity which dealers/market makers help facilitate. Market liquidity is a result of funding liquidity (they both impact each other as assets are used as collateral). The Fed acted as a dealer of last resort during the crisis, and has always acted as one in t-securities. It wasn’t surprising that in a time of crisis, the notion of fungibility was impaired as markets broke down.
Trading securities is not the same thing as trading 2 $5 dollar bills for 1 $10 bill. Securities are different in kind whereas the bills are not. Both bills are the liability (if physical notes) of the same issuer, the Fed. When trading securities this most likely is not the case as the securities swapped were likely issued by different counter-parties.
“the price of a new issues will depend primarily on the issuer’s ability to service the debt or generate returns on equity.”
The price you are describing is a theoretical “fundamental” price which may or may not accurately be reflected in the actual/realized market price we see. The price we see is based on bid/asks which is a product of funding and market liquidity. The Fed can affect both funding and market liquidity.
Isn’t this the point? The Fed via OMO and expectations can impact asset prices of specific assets (t-securities and agency MBS in QE3) which then permeates throughout the capital markets based on the pattern of trading. Monetary policy works partially due to its ability to affect the relative price of financial securities relative to goods/service prices. The hope is that the relative price rise in financial assets will induce holders of securities to start consuming goods/services at a greater rate. This affects AD. Rising asset prices also affects expectations of AD. This is especially true when considering equity prices which are not the liability of the issuer (in the same way debt is). The Fed can increase financial net worth in the system. However, the distribution of that net worth is skewed toward those holding the most financial assets. So I respectfully disagree, it certainly does matter who is holding the existing paper at any given time. I’d be ok with this policy channel if financial assets were more equally owned in our society, but they are not.
“If you have a mortgage, the Fed might already own your securities.”
If the Fed were transparent in who they were going to buy securities from (as they are in QE when stating they will buy T-securities and agency MBS), the issuer could essentially set the terms*. This becomes even more important if the Fed decides to increase the monetary base permanently by never selling the specific securities. The issuer would be able to perpetually roll it over.
*If the Fed made it clear to me and the market that they would be buying my loan, I’d set the terms to 0% interest where the entire principal would be paid on 2114.
17. January 2014 at 10:04
Rafael, I think you misunderstood me. I agree there will be no hyperinflation, but only because the Fed will sell securities or raise IOR when it needs to. If it did a helicopter drop it would lack the resources to do so.
dtoh, Shifts in money demand.
Vaidas, I don’t see where 1987 created any moral hazard. A better argument is that 1998 did.
libertaer, You can’t get something for nothing. If the government gives people money there is an opportunity cost, higher future taxes. There’s no escaping that fact. If you try to print money to pay off future IOR liabilities, you’ll end up with hyperinflation.
17. January 2014 at 15:12
I think the miunderstanding results from a miscommunication when defining helicopter drops.
Congress would instruct the treasury to issue t-securities and then send out equal checks to every citizen. Using their discretion, the Fed would then acquire the t-securities or not. If they do, both bank deposits/broad money and reserves/base money would increase by equal amounts. No bureaucratic decision making, pure direct transfers. This would have knock-on effects which would depend on how the broad money balances are used. Thus, the Fed would have the resources to drain reserves if needed depending on their policy targets.
As for getting something for nothing, I don’t necessarily disagree. The new broad money balances would either be used to bid up/acquire assets which would change their relative price compared to goods/services, used to buy goods/services, used to pay off debt (a form of buying back their own assets), or accumulate for later use.
As economic activity/transactions and real output increase, tax receipts will endogenously rise (as we are partially seeing now). If the economic activity starts to overheat as there are less idle resources and lower unemployment, sellers would begin to increase prices for goods/services as their ability to increase output is constrained. The output gap would be closed. At this point, the Treasury would use the tax receipts collected to buy back t-securities held by the private sector. In effect, the Fed would sell back the treasury securities which it could have accumulated during the heli-drops. This would drain the MB. The treasury would buy back the treasury securities, which would drain them as well.
Thus, the higher taxes (relative to the counter-factual without heli-drops) come at a point where they are logically needed. Heli-drops/direct transfers are essentially tax cuts, but tax cuts which are distributed equally to all citizens making it a more neutral policy. There would be no need for a slow bureaucratic decision making process. The heli-drops could start small and increase incrementally depending on how they impact economic activity.
The Fed may be able to offset inadequate fiscal policy, but it doesn’t excuse Congress from doing their jobs poorly. Unlike the Fed, they can directly affect the distributional outcomes. Direct equal transfers to everyone has the least distributional effects.
17. January 2014 at 16:27
Rafael,
Got it. But IMHO that’s just fiscal policy especially if the Fed is issuing securities.
1. Politically not going to happen.
2. It would work in theory.
3. It’s highly distributional. Unless the Fed is just printing money, someone has to pay for it.
17. January 2014 at 16:28
Rafael,
Should readd if the “Tsy is issuing securities”
17. January 2014 at 16:30
Scott,
“Shifts in the demand for money.” Yes… if unpredictable, but if we ignore ER, the as a practical matter shifts are only related to short term rates and these shifts are pretty predictable.
17. January 2014 at 17:07
1. Agree.
2. I agree I think it does work in theory.
3. Reread the parts having to do with taxation occurring endogenously as transactions expand due to an increase in economic activity (which is again partially happening now). If economic activity is overheating as described, taxes receipts generated would be used to buy back and retire t-securities at a point when it’s logical to do so. If needed or desired an additional tax can be applied to everyone in the same $ amount just as the heli-drops were transferred to everyone in equal $ amounts.
Compare the above relative to what Congress is already doing where the budget deficit and treasury issuance are used in highly distributional ways. Heli-drops are more efficient and transparent relative to having Congress make tax-spend decisions that result in the budget deficit/treasury issuance outcomes. If one supports larger budget deficits to finance tax cuts, as I do, then heli-drops are a superior form as they have less of a distributional outcome for reasons described in the prior post.
From my understanding the Fed sets monetary policy to hit its policy goals regardless of how inefficiently Congress uses the budget/treasury issuance process. As long as the Fed can only acquire T-securities, Congress is the only agent of the government which can directly impact distributional outcomes. Just because the Fed has the ability to offset the overall effects of poor fiscal policy, that doesn’t mean they can account for the distributional outcomes that derive from poor fiscal policy.
Going back to your #1 point it is politically not going to happen because heli-drops reduce the need for Congress to make allocative decisions. Why would they ever agree to the distribution of equal direct transfers to all citizens, even if its a small part of the budget deficit, as it would allow the recipients to chose how they are used, not Congress.
Like Scott recently stated “Remember that Congressmen are the lunatics and Fed officials are the “grownups” in the policymaking realm.” I couldn’t agree more and poor fiscal policy is causing certain unequal distributional outcomes which Bernanke has stated that the Fed cannot overcome. Reduce the dependence on Congress. Direct transfers is a good way to start.
18. January 2014 at 09:34
“libertaer, You can’t get something for nothing. If the government gives people money there is an opportunity cost, higher future taxes. There’s no escaping that fact. If you try to print money to pay off future IOR liabilities, you’ll end up with hyperinflation.”
Either we’re talking past each other or somebody is making a mistake, which must be me ’cause you’re the expert here.
For me, you’re sounding suddenly like a backing theorist, but I know that you’re into quantity theory. Hm..
I think we both agree that instead of selling treasuries the central bank could pay interest on reserves to fight inflation.
Now, why do you think these interest payments have to come out of taxes? And why hyperinflation? You only get interest if you lend base money to the CB, which is then taken out of circulation. If the interest rate on base money is 2%, you have to lend 100$ to the CB to get 2$ in interest. So base money in circulation goes down 98$. Base money which just sits at the CB can’t create inflation.
18. January 2014 at 16:21
Rafael,
When the economy picks up, the tax revenue increase does not come equally from everyone. Everyone gets the distribution, but not everyone pays taxes.
Also as I have said, I don’t believe normal OMP are distributional (or at least they are not distributional to any meaningful extent).
An additional tax amount can be applied to everyone in the same $ amount as the heli drop. I’m sure you can give money to the panhandler in front of my apartment, but I’m just as sure you can not get it back.
18. January 2014 at 17:39
If you come to the determination that their exists an output gap that is sizable where individuals are unemployed and underemployed, a helicopter drop can close the output gap by increasing real output. The closing of the output gap in theory should result in a greater number of individuals gaining employment who end up paying taxes. As the labor market tightens, it would be rational to assume that nominal wages could start to rise especially as the underemployed are fully utilized.
If you favor deficit financed tax cuts to those currently employed, helicopter drops are fairer in that those out of the labor force also get the same distribution. Tax cuts end up getting paid at some future date by some individuals who never directly received a distribution. This would include those currently unemployed who end up gaining employment and paying taxes as the output gap closes.
So yes when the economy picks up, I agree that the tax revenue increases do not come equally to everyone. Surely the unemployed won’t mind the trade off of receiving a distribution now (when they are most likely cash and credit constrained) in exchange for paying taxes when they gain employment. I’m sure the underemployed who are able to find more appropriate and higher paying jobs as the economy picks up won’t mind paying higher taxes on their higher incomes either.
As for monetary policy if the idea is that LSAP’s work by exchanging more liquid assets for less liquid assets which results in greater funding and market liquidity, this has shown to create financial conditions which coincide with rising asset prices. At the very least, this leaves those holding the highest portions of assets better off relative to those who don’t. As financial prices rise especially relative to goods/service prices, this should result in the marginal increase in exchange of financial assets (by those holding them) for goods/services.
18. January 2014 at 18:01
Slight correction to the above
*”This would include those currently unemployed who end up gaining employment and paying taxes as the output gap closes. In effect they end up paying taxes after gaining employment but do not receive the full benefits of the transfers which result from the tax cut while they are unemployed.”
19. January 2014 at 02:46
Rafael,
Sorry to keep disagreeing, but
1. The bulk of any taxes will be paid by a small percentage of the top producers (i.e. highest income) in the economy. The distribution you propose is in equal amount. Sure, for some people the money received may be roughly equal to the tax later paid, but unless this is true for everyone then the proposal is definitely redistributive.
2. As I have noted, successful Fed action will typically raise NGDP expectation leading to a higher Wicksellian rate and higher bond yields (lower prices) so it is not possible to generalize that those holding the highest portion of assets will benefit as this depends on their portfolio. Those holding equity portfolios will do well. Those holding fixed income portfolios will be worse off. And as I have pointed out before, those holding assets will generally have any gains or losses offset by the opposite gains or losses when growth began to decline.
19. January 2014 at 07:40
1. Fiscal policy matters partially because it produces a budget deficit which triggers t-security issuance. The Fed can offset poor fiscal policy by using monetary policy, but this to a large degree still involves buying t-securities. If they purchase privately issued securities especially in a transparent manner, they change the relative price of the security which makes the private issuer better off relative to other securities issuers (we will just have to disagree if you don’t think this would give this entity an advantage). Imagine if congress attempted to run massive surpluses in order to pay off all the debt, the total supply of treasuries would begin to decline in a way where the Fed could be forced into buying private securities. Secondly, my guess is that there would be an AD shock which the Fed may not be able to totally overcome. Thus monetary policy ALWAYS matters (even at the ZLB), but so does fiscal policy.
2. Since a budget deficit is needed (and most likely a much larger deficit right now), then the point of the discussion is to think about which forms of FP are the least distributional. One discretionary form of increased FP would be congress deciding how to spend. They would issue treasuries and credit the proceeds to those connected to their decisions. The future taxes are certainly paid by many who did not directly receive the distribution or will benefit from Congress’ decision. Another form of FP is tax cuts, but as mentioned above heli-drops are less distributional.
Also, given that some t-securities continue to exist (either held by private sector or Fed), it is important to appreciate that some FP never gets “paid off” as the private sector and Fed are willing to hold t-securities. As long as there is a government debt balance outstanding, some amount of t-securities are continuously rolled over. Thus, no future taxes ever result.
As for any taxes that are paid in the future, it is true that the top earners pay the most individually in terms of total $ amounts, but it might not be true that they pay the most as a percentage of their income or financial net worth. My guess is that those that gain employment and the underemployed that see their wages increase will end up paying a higher % (paying minimal taxes when unemployed to paying taxes when employed is pretty big % change in taxes paid).
2. From my understanding one of the goals of LSAP’s is to raise financial asset prices and thereby the total financial net worth of asset holders on aggregate. In the event that the Fed is able to successfully raise NGDP and this leads to higher bond yields, as you state those holding the equity shares are better off relative to those holding bonds. Even here my guess is that in order for the Fed to successfully raise NGDP via inducing a higher marginal exchange between financial assets and goods/services, it would require that the gains in equities be higher then the potential losses on bonds. So an individual holding an equal mix of shares and bonds should be made better off overall despite potential losses on the bond portion of their portfolio (in large part, this is what we have seen since 2009).
If the gains made by those holding equity shares are offset by the losses made by those holding bonds, then equity holders are made better off relative to everyone else. Since shares exist in finite quantity at any given time, someone will have to experience the gains while others bear the losses. Not everyone benefits equally from higher equity prices if the bulk of share quantities are concentrated in the hands of a relatively small percentage of individuals.
In your last sentence, I think you are describing a scenario where the Fed fails in targeting a trend growth. Also, it is absolutely very far from certain that IF growth slows that some who have been holding assets will have all their gains evaporate. Some will, some won’t, and some who didn’t experience the gains will experience the losses nonetheless (timing and trading play a huge role, path dependence). That said, if history is any indication (which it very well might not be) then in the long run financial assets holders on aggregate will be better off because their gains will not be offset by future losses in the even of a growth slow down. Someone who bought an array of financial assets in say the 1980s and held them until today has seen their asset holdings increase in price despite 2008. Even if they sold all their assets at the very bottom of 2009, they are most likely still better off.
Bernanke recently spoke about some of the topics above. His conversation can be found at the link provided. The whole thing is worth listening to but definitely see how he responds to the question beginning at minute 45.
http://www.brookings.edu/events/2014/01/16-central-banking-after-the-great-recession-bernanke#/full-event/
19. January 2014 at 07:43
dtoh,
“Central Banking after the Great Recession – A Conversation with Ben Bernanke” can be found by clicking on the Full Event tab found on the right side of the video. If you scroll down it is at the bottom.
http://www.brookings.edu/events/2014/01/16-central-banking-after-the-great-recession-bernanke#/full-event/
19. January 2014 at 07:59
This may also interest you.
https://www.youtube.com/watch?v=ZhrY_coLK_k
19. January 2014 at 18:48
Rafael,
Looked at the video thanks. On your specific points.
1. Purchase of private assets does not confer a benefit to one or a few particular issuers. Assets are fungible and there will be no significant change in relative prices. Anyone seeking to issue new securities or exchange existing securities for money will benefit equally.
2. I don’t think we are anywhere near running out of government securities.
3. ZLB is entirely an illusory problem.
4. Fiscal policy is effective in raising AD, but much less effective than monetary policy and IMHO often leads to long term structural impediments to growth.
5. Everyone is better off when the economy grows more rapidly. Asset holders will probably benefit more than non-asset holders. IMHO distribution of wealth and income are relatively unimportant. We should be concerned instead with consumption differences.
6. There are always winners and losers in a market. That’s why the system works. To the extent that we have stable growth there are fewer surprises and greater predictability. IMHO this leads to a fairer systems where winners and losers are determined by skill at allocating assets rather than by incompetence on the part of policy makers and politicians.
7. Equity holders have enjoyed gains not enjoyed by others recently, but they also got creamed 5 years ago.
8. In summary, OMO makes virtually everyone better off. Clearly some will do better than others, but it is a fair, transparent, and effective approach. OTH, helicopter drops are highly redistributive and thus are both unfair (IMHO) and will reduce the underlying long term growth prospects for the economy.
19. January 2014 at 21:34
dtoh,
We will just have to agree to disagree on these topics. I’m guessing our different conclusions have a lot to do with our backgrounds. The mental model I use to understand the macro/monetary and financial systems more closely resembles the work of Perry Mehrling, Adair Turner, George Soros, Bernanke, FT Alphaville, Katherina Pastor, Ashwin Parameswaran, Edward Harrison, Frances Coppola, among others (with a Post-Keynesian bend). Trend following and technical trading also have taught me some invaluable lessons.
I’ve learned and had to adjust my thinking a lot since the crisis period and that will probably be the case forever given the complexity involved in these affairs. It should be exciting and incredibly interesting to see how these coming years unfold. There will no doubt be a lot to absorb. It will be interesting to see how this conversation holds up.
Again, thanks for the chat. I appreciate the time you took to respond.
If you haven’t had the chance to check out the Adair Turner presentation linked above, definitely take a look. Katherina Pastor’s paper “Towards a Legal Theory of Finance” is also fascinating.
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2178066
19. January 2014 at 22:36
Rafael,
Same here. I’ll have a look at the Turner presentation.
17. September 2019 at 00:47
[…] to try to reincarnate all those arguments: I recommend reading Sumner in particular, pieces like this or this. But I am also very influenced by two other things. First, there is the record of Japan […]