Why it’s difficult to give advice on monetary policy tactics
In a sensible world it would be relatively easy to work through monetary policy options. The first step would be to ask where the central bank wants to go. Everything else would hinge on that strategic decision. Thus if they want 5% expected NGDP growth, level targeting, you can pretty easily work out some tactics to achieve that target. One such policy would be to have the NGDP futures market determine the monetary base and the fed funds rate.
Under that sort of policy regime there’d be much less reason to worry about a collapse of a major bank. That’s partly because the NGDP target would help prevent the bank failure from affecting output, but much more importantly it would make it much less likely that a major bank failure would have a contagion effect. Because bank bailouts are really, really, undesirable, that’s a huge advantage of NGDP level targeting. Even so, this doesn’t mean the Fed necessarily made a mistake in bailing out the banking system in late 2008. After all, the Fed does not engage in level targeting, at least when the nominal interest rate falls to zero. So it’s hard to be certain that a banking crisis would not lead to lower NGDP. And in a world of sticky wages, falling NGDP is a really, really, really bad thing (that’s one more ‘really’ than for bank bailouts.) Given that policymakers are risk averse, I’m not surprised that they didn’t want to take any chances.
Cardiff Garcia has a long article on IOR that indirectly illustrates this problem. A sensible central bank would do the following:
1. First decide where it wanted to go (i.e. set an NGDP target.)
2. Then provide the level of base money (and hence the fed funds rate) that is expected to hit that target. The Lars Svensson approach.
3. If the equilibrium fed funds rate is near zero when expected NGDP growth is on target (not likely), then decide whether that level of rates is likely to bankrupt the money market mutual fund industry, and also whether that sort of bankruptcy is systemically important. (An issue I’m not qualified to address.) If so, set an IOR at 0.25%, and recalculate the level of base money required to produce on-target expected NGDP growth.
4. Then radically reform the MMMF industry so that it can never again hold our monetary policy hostage.
Unfortunately our policymakers only do step 3, and ignore steps 1, 2 and 4. That’s why Garcia has such a hard time sorting through the options, (although he, Yichuan Wang, David Beckworth, et al, do an excellent job of trying.)
This column in Free Exchange shows how badly we botched the regulation of MMMFs:
After the crisis was over, the money markets were an area that seemed worthy of further scrutiny. Like asset-backed securities before them, they had appeared boring and safe when they were actually a large source of hidden risk. The industry, however, put up an unprecedented lobbying campaign against the SEC. As a result, it was able to avoid anychanges the existing rules, even though they had proven so hopelessly inadequate prior to the crisis. How could this have happened? An excellent and detailed investigative report by Bloomberg provides an answer: the SEC is a captured organization whose employees regularly alternate between government and lobbying on behalf of the firms the SEC is meant to keep in line. It is not right to call this a “revolving door,” since that implies some kind pause while going from one side to the other.
Let’s be clear. If the events described in the Bloomberg article happened in China, everyone would immediately (and rightly) decry cronyism and corruption. Why not do the same here?
This is just the tip of the iceberg. After the 2008 financial fiasco there were lots of calls for “re-regulation,” especially by the Democrats. Then in early 2009 they seized the House, Senate and the Presidency. They needed to address 4 basic weaknesses in our financial system:
1. A deeply flawed FDIC system that allowed (mostly smaller) banks to lend insured deposits to developers undertaking highly risky projects.
2. A deeply flawed FDIC system that allowed (mostly smaller) banks to lend insured deposits to people taking out subprime mortgages, i.e. mortgages with less that 20% down.
3. Fannie and Freddie, which needed to be blown up.
4. The MMMF industry.
And in its infinite wisdom the Congress succeeded in writing a 1000 page bill that addressed none of these key problems. Which means the crisis is likely to happen all over again. Just as the post-1980s S&L fiasco “re-regulation” solved nothing. As far as Congress is concerned moral hazard isn’t a bug, it’s a feature. Of course things are no different in Europe or China, so we shouldn’t be shocked by any of this. And I doubt the GOP would have done any better in addressing these issues.
It’s a messy world, which makes it hard to evaluate monetary policy tactics. On the other hand NGDP level targeting would clarify a lot of things. The biggest advantage may not be better macroeconomic stabilization, but rather better policy in other areas. Just as the Great Moderation made the world safe for neoliberalism, NGDP targeting would make the world safe for sensible financial regulation.
PS. It’s frustrating to see people focusing on the effect of lower IOR on lending. The goal isn’t more lending; it’s a lower demand for base money.
Tags:
10. September 2012 at 21:32
I am a non-economist but I am keenly interested in these issues and trying to understand NGDP targeting. Here is what I do not understand. Suppose the Fed decides to target NGDP growth of 5%. Say NGDP is actually running more like -0.5% and IOR is 0.25% (i.e., we are up against the ZLB). What is the Fed supposed to *do* to push NGDP growth to 5%? Is it really sufficient for people to believe the Fed wants it to be 5% and will keep interest rates low even once inflation picks up until average NGDP growth over some k-year period hits the magic 5% level? i.e., is there anything more than people’s future expectations at work here? How is that enough? How will that make those who are hoarding money now (China, the wealthy, corporations) and investing in 10-yr treasuries with negative real returns suddenly decide to spend? Getting people with money to decide to spend it is what boosting AD is all about, right? I don’t see how NGDP targeting leads rich people to buy more yachts or hire more servants, which is basically what’s required for households to deleverage. (Well, that or public sector action e.g. fiscal stimulus, but that seems politically impossible just now.)
I’d really appreciate any response.
Kenneth Duda
kjd@duda.org
10. September 2012 at 21:52
Except there is no nexus between proposed SEC regulations and your point 4: “Then radically reform the MMMF industry so that it can never again hold our monetary policy hostage.”
http://www.federalreserve.gov/releases/cp/rates.htm
There is plenty of paper to buy at positive yields. Existing rules require these funds to hold investment grade paper. Yes, that doesn’t mean riskless. What’s the problem here exactly?
10. September 2012 at 22:25
Scott,
“1. First decide where it wanted to go (i.e. set an NGDP target.)”
This part here is what every consultant in every industry knows or should know.
It’s one of the lessons that stuck when I visited and I was told what consultancy was about: first you talk to your client to figure out where they want to go and why they want to hire you. Often the client does not even realize that they don’t have a goal in mind and that’s why that first step is so crucial. Don’t ask that question and you’ll have failure.
10. September 2012 at 22:45
Isn’t lower demand for base money the same thing as more lending in the present scenario? If banks the demand for base money results in large amounts of money parked in excess reserves and lower IOR results in a lower demand for base money, how else can banks reduce their holdings of excess reserves besides through lending?
11. September 2012 at 00:26
If you change ‘base money’ to ‘money, properly defined’, I will agree with you! Whether or not negative IoR lowers long rates, as long as it destroys some risk averse capital, it will be an unmitigated good for the economy.
The people at FT Alphaville are very good, but they have an unhealthy and schizophrenic obsession with the health of the financial industry, as it exists today. You could destroy large parts of the industry without losing much for the real economy.
People forget that when money markets seized up and financial credit spreads skyrocketed in the immediate wake of Lehman’s crash, non-financial commercial paper was still going strong.
11. September 2012 at 03:14
John:
Banks can rid themselves of excess reserves by purchasing bonds. While this is lending broadly understood, it isn’t making new commercial or consumer loans.
11. September 2012 at 03:58
ssumner:
1. First decide where it wanted to go (i.e. set an NGDP target.)
2. Then provide the level of base money (and hence the fed funds rate) that is expected to hit that target. The Lars Svensson approach.
3. If the equilibrium fed funds rate is near zero when expected NGDP growth is on target (not likely), then decide whether that level of rates is likely to bankrupt the money market mutual fund industry, and also whether that sort of bankruptcy is systemically important. (An issue I’m not qualified to address.) If so, set an IOR at 0.25%, and recalculate the level of base money required to produce on-target expected NGDP growth.
4. Then radically reform the MMMF industry so that it can never again hold our monetary policy hostage.
Unfortunately our policymakers only do step 3, and ignore steps 1, 2 and 4.
No, “our” policymakers do all these steps.
They do 1. via their consumer price inflation target, which is 2-3% per year in the long run. The PCEPI has been approximately 2% average per year since January 2000.
They do 2 via 1.
They do 4. In fact, that have been doing 4 for decades. Tinkering and tinkering.
11. September 2012 at 04:38
Kenneth, No, you are not trying to get those with money to “spend” in the ordinary sense of the term. When you save by buying assets, you are “spending” in a monetary sense.
Jon, It’s the “break the buck problem.” The shares should not be priced at $1.
John, They can buy assets like bonds.
11. September 2012 at 04:58
Scott, I have followed your blog intermittently for several years. Correct me if I’m wrong, but I don’t remember your advocating MMMF reform in 2009 and 2010. What stands out in my mind from that period is your position that the financial crisis didn’t cause the recession, and that those of us who believed there to be a causal relationship between the two were mistaken.
Here, on the other hand, you seem to be saying that, conditional on the central bank *not* doing NGDP targeting, bank failures etc. can “affect output” and “lead to lower NGDP.” So I remain curious whether you think the financial crisis that began in Aug. 2007 and continued through early 2009 had a significant effect on output. My answer has always been “yes,” and I’ve always interpreted you to say “no.”
I think your short answer would be that the central bank controls the nominal economy, and if the Fed *had* been doing NGDP targeting, there would have been no recession. So the causal story that says the financial crisis affected output is conditioned on a flawed monetary policy regime.
Yet in this post, you say that *even if* the Fed adopts NGDP targeting, it should still do “radical” MMMF reform. But why? If the Fed is doing NGDP targeting then a run on the MMMF industry can have no effect on output, under your view as I understand it.
11. September 2012 at 05:04
How the heck is buying bonds not making a loan? You’re loaning your money to an entity at interest for a specified time period. That sounds pretty much like a loan to me. What is a home loan, the bank lends money to a homebuyer at interest for a specified time period.
11. September 2012 at 05:31
Your list of four problems with the financial industry doesn’t touch the most important items.
The primary channel for the excessive mortgage credit was securitized CMOs, and those were most definitely not where the (mostly smaller) banks were playing.
Giant investment banks levered 30 to 1, synthetic securities, conflicted rating agencies and credit default insurance with no reserve, underwriting or disclosure requirements were much greater problems than the four you listed. The book “The Big Short” and articles about the Magnetar trade and the Guassian copula function are highly illuminating.
Money market funds breaking the buck were merely a symptom of the excessive leverage of the big investment banks and the lack of competence and independence of the rating agencies.
11. September 2012 at 07:43
FWIW, but George Soros is now advocating NGDP targeting. At the end of the piece:
http://www.project-syndicate.org/commentary/why-germany-should-lead-or-leave-by-george-soros
11. September 2012 at 08:01
‘Even so, this doesn’t mean the Fed necessarily made a mistake in bailing out the banking system in late 2008.’
Okay, good. You had me worried there for awhile.
Analogously, just because a physician decides he needs to operate to save your life doesn’t mean he approves of your lifestyle–smoking, drinking, overeating…–that put you into the ‘crisis’.
11. September 2012 at 08:38
This paper argues that SEC employees act aggressive in order to get hired when they leave. The authors refer to it as the “human capital hypothesis” in contrast to a “rent seeking hypothesis”:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2125560
11. September 2012 at 09:28
MR, There are several issues here:
1. I’ve favored financial reform for decades, due to the moral hazard problem created by bailouts.
2. If the Fed is doing its job, financial distress won’t have much impact on output (maybe a little due to “re-allocation.”)
3. My preference for MMMF reform is mostly based on the assumption that monetary policy will not be reformed. You are right that this post didn’t make that clear. If it is reformed, then there’s no great need for MMMF reform. But we’d still need to reform banking, the GSEs, etc, for moral hazard reasons.
John, I agree. But most people define loans to exclude buying bonds. As when they ask “what if the banks can’t find anyone to lend to?” Or “what if there are no good credit risks?”
11. September 2012 at 18:42
Does it not seem that wages are even stickier now than they used to be? (I suspect there is solid evidence for this, though I don’t know what it is.) One kind of structural reform that would help us out of the present recession, even without good monetary policy from the Fed, is the kind that made wages more flexible. A good start would be the repeal of most federal labor legislation. Of course, that seems politically unrealistic, but today’s unrealistic proposal might possibly be tomorrow’s conventional wisdom.
Continued bad policy by the Fed makes structural reforms all the more important. Granted, the politically realistic view is that we will continue to have bad monetary policy and no reforms to promote flexibility.
11. September 2012 at 20:05
“have the NGDP futures market determine the monetary base and the fed funds rate”
Using futures as a “guide post” won’t work. History has shown that neither economists nor speculators know how to forecast. And the proverbial “monetary base” is not a base for the expansion of new money & credit, e.g., currency held by the non-bank public would induce bank credit contraction if not offset by (e.g., open market operations of the buying type). And excess reserves currently remunerated at .25% absorb existing bank deposits within the CB system & attract voluntary savings from the non-banks (inducing dis-intermediation within the financial intermediaries).
Note that using Dr. Daniel L Thornton’s required reserves as a “base” changes the system’s “expansion coefficient”.
See: http://bit.ly/yUdRIZ
Quantitative Easing and Money Growth:
Potential for Higher Inflation?
Daniel L. Thornton
“the close relationship between the growth rates of
required reserves & total checkable deposits reflects the
fact that reserves requirements apply only to checkable
deposits”
This multiplier (ratio of reservable liabilities to required reserves) would have to be adjusted to account for structural changes that would alter the trend ratio of time/savings accounts to transaction based accounts, etc.
In the long runthe trend of average prices & interest rates reflect the actual relationships between rates of change in monetary flows & rates of change in real-gDp.
Other considerations: the only tool at the disposal of the monetary authority in a free capitalistic system through which the volume of money can be controlled is via legal reserves & reserve ratios. The first rule of reserves & reserve ratios is to require that all money creating institutions have the same legal reserve requirements, both as to types of assets eligible for reserves, as well as the level of reserve ratios. Monetary policy should require that the commercial banks have UNIFORM reserve ratios, for ALL deposits, in ALL banks, irrespective of size.
In addition, the only type of bank asset that Fed can constantly monitor & absolutely control, are inter-bank demand deposits (IBDDs) held in the Reserve banks, owned by the member commercial banks (like the ECB). This was the original definition of legal reserves in 1913, and it is still the only viable definition (pre-1959 requirements pertaining to assets). Thereupon the Fed can use contemporaneous reserve requirements & not a reserve maintenance cycle where the computation period is 14 days followed by a 17 day lag & finally ending with the 14 maintenance period.
Monetary policy objectives should be formulated in terms of desired roc’s in monetary flows (MVt) relative to roc’s in real-gDp. Contrary to economic theory, & Nobel laureate, Dr. Milton Friedman, monetary lags are not “long & variable”. The lags for monetary flows (MVt), i.e. the proxies for (1) real-growth, & for (2) inflation indices (for the last 97 years), have been mathematical constants. However, the FED’s target (interest rates), is INDIRECT, varies WIDELY OVER TIME, & in MAGNITUDE.
And roc’s in bank debits (our means of payment money times its rate of turnover-MVt) can serve as a proxy for all transactions (aggregate monetary purchasing power). I.e., allow member bank legal reserves to grow at no greater rate than would allow rate of increase in monetary flows to equal the rate of increase in real output. There is evidence to prove that rates of change in nominal gDp can serve as a proxy figure for rates of change in all transactions. Rates of change in real-gDp have to be used as a policy standard.