Friday, August 3, 2012

But why, JW? Why "The past 25 years"??


I think of finance as a competitor of the Federal Reserve, when it comes to providing funds to the economy. It follows that the growth of finance, the relative size of finance, and the deregulation of finance play a role in the outcome of that competition.

At The Slack Wire, JW Mason writes:
Textbook macro models, including the IS-LM that Krugman is so fond of, feature a single interest rate, set by the Federal Reserve. The actual existence of many different interest rates in real economies is hand-waved away with "risk premia" -- market rates are just equal to "the" interest rate plus a premium for the expected probability of default of that particular borrower. Since the risk premia depend on real factors, they should be reasonably stable, or at least independent of monetary policy. So when the Fed Funds rate goes up or down, the whole rate structure should go up and down with it. In which case, speaking of "the" interest rate as set by the central bank is a reasonable short hand.

How's that hold up in practice? Let's see:


The figure above shows the Federal Funds rate and various market rates over the past 25 years. Notice how every time the Fed changes its policy rate (the heavy black line) the market rates move right along with it?

Yeah, not so much.

In the two years after June 2007, the Fed lowered its rate by a full five points. In this same period, the rate on Aaa bonds fell by less 0.2 points, and rates for Baa and state and local bonds actually rose.

I tried to duplicate Mason's graph. But I had to guess from a large selection of interest rates. Here's what I came up with. It goes back to the same start-date:

Graph #2: Collected Interest Rates (since 1987)
Eh, it's a pretty good match. (Note that I am using TB3MS, my "proxy" FedFunds series which goes back 20 years beyond FEDFUNDS. (FEDFUNDS is still on the graph, pretty well hidden by the purple line.))

Here's a look at the numbers as far back as they go:

Graph #3: Collected Interest Rates
Click for FRED source page

What I want to do is
1. drop FEDFUNDS from the data, and use the proxy data series.
2. set the proxy series apart, and average all the rest together.
3. compare the FEDFUNDS proxy to the average of the other rates.

JW Mason says market rates do not move right along with the rate set by the Federal Reserve. That is true, if we go by his graph. But his graph only goes back to 1987, and our economy was already well-financialized by then.

I think if we look farther back, to a time when we were less financialized, the graph may show that markets DID move with the Fed rate. Anyway, that's my guess.


Here's what I did:
1. Download the data in the graph, from FRED.
2. Work on the data using Open Office Calc.
3. Eliminate the early years, where there is no FedFunds Proxy for comparison.
4. Reorganize the FRED source columns: move data series with later start-dates farther to the right. (This is for my convenience in selecting cell ranges when figuring averages in Excel.)

This is a lot of data. All monthly, it is 12 times what I'm comfortable working with.

Note: The start-dates of the various interest rates differ. The corporate AAA and BAA begin well before 1934, when the FedFunds Proxy begins. MSLB20 begins in 1953Q1. GS10 begins 1953Q4. And MORTG begins in 1971Q4. The number of interest rates in the average increases as each new series begins.

Hm. I used Open Office Calc to get the size of the graph. Then I went into Paint and set the size of a new, blank picture to the same size. Then I copied and pasted. The vertical bar thickness changed. Several of the vertical bars from Calc did not show up in Paint. And the colors changed.

Graph #4: Wow, that's ugly!

Used the Windows "snipping tool" on the Calc graph, and got this version:

Graph #5: The FEDFUNDS Rate versus an Average of Other Rates

Well that's better.

I don't know if my expectation holds up. It doesn't really look as if market rates moved with the Fed rate, even before the Great Financialization. Actually, it looks like market rates are "sticky down".

The other-rates average (red) is definitely higher than the FedFunds proxy in the two Depression-times, when the FedFunds rate is near zero. And it looks to me that the red and blue are much closer on the up-swing (1946-1982) than on the downswing (since 1982). That could just be because "other" rates are not immediately responsive to changes in the FedFunds rate. (Long term, obviously, they rise and fall together.)

But maybe the two are closer on the rise than on the fall because financialization was lower on the rise than at other times. You can gage the level of financialization on Thomas Philippon's graph and mine, both presented here.  The striking fall in financialization shown on those graphs reaches a bottom in the mid-1940s, just when the up-swing begins on Graph #5.

I don't have the math skills to evaluate the result any better than that. But I can say that this evaluation of interest rates does not contradict my view that private finance competes with the Federal Reserve and that when finance grows large, it circumvents and weakens policy.

9 comments:

Unknown said...

I've also wondered about this question, which seems particularly critical for Europe today. As it relates to the US though, here is a recent comment from Randy Wray (http://bit.ly/RhqUAJ):
"Forget about market determination of rates on treasuries. It’s not about supply and demand. The monopoly issuer of the currency determines the overnight rate. QE then drives the rate to the central bank’s “support rate” (rate it pays on reserves), and if the central bank’s commitment to low rates is credible, the term structure adjusts downward. It is hard to drive long rates below 2%, as Keynes argued, due to the risk of capital losses (wiping out the interest earned).

And if the markets don’t believe the central bank is committed to ZIRP it can take a while to bring down the long rates—however reality eventually dominates expectations so a determined central bank will bring down rates even on long term treasuries (as the Fed eventually did). Funny how expectations converge to reality rather than the other way around as Bernanke and other “New Monetary Consensus” economists believe."

My initial impression is that this might help explain the path of other rates as well. Banks/markets are slow reacting initially to Fed action since the next step and length of directional movements is unclear. As markets become more certain of Fed policy going forward, interest rates shift to match updated expectations/reality.

Jazzbumpa said...

my view that private finance competes with the Federal Reserve and that when finance grows large, it circumvents and weakens policy.

You have two propositions here: 1) private finance competes with the Federal Reserve; and 2) when finance grows large, it circumvents and weakens policy.

It is not obvious to me that there is any connection between the two. I absolutely agree with the latter - as you well know.

I'm not so sure about the former. It's seems like the flip side of "crowding out" - an idea that conservative economists love, and I think is pretty much bull shit.

I agree with Joshua about expectations and reality. But not about the power of central banks. They follow, not lead the market.

http://www.angrybearblog.com/2012/05/who-determines-short-term-interest.html

Cheers!
JzB

Unknown said...

@Jazz

I'm not sure that post at Angry Bear contradicts my claim, supported by Wray and others. The Fed acts in certain intervals and operates under a corridor system:
http://feedproxy.google.com/~r/LibertyStreetEconomics/~3/GxrkVSocRYE/corridors-and-floors-in-monetary-policy.html
In this manner, the Fed sets the target rate but permits fluctuations within a band between the discount rate and IOR rate. T-bills will take into account upcoming policy changes by the Fed and may therefore lead actual policy (especially if policy changes are transparent).

So I agree with Angry Bear, but don't think it implies that the Fed is powerless in controlling the rate on Treasuries. I should also clarify that this power extends less to market interest rates and even less to the extension of credit by banks.

Jazzbumpa said...

Joshua -

You're making an expectations-based claim, where Fed influence drives a market determined rate into their target range. I don't think there is any way to either prove or disprove it.

From your link:
The Federal Reserve conducts monetary policy by choosing a target for the federal funds rate, which is the average, market-determined interest rate at which banks and certain other institutions lend funds to each other on an overnight basis.

So the idea that the FFR is market-determined is recognized by the Fed.

But my point is removed one step. I'm saying the target is a reaction to market forces, not a leader of them. As evidence, I note that the FFR always follows and never leads the market-determined 3 mo T-Bill rate.

I realize this is a minority, and in fact contrarian, PoV, and that many people are offended by it. But it is totally in agreement with the data.

Cheers!
JzB

Unknown said...

JzB,

Fair point about my claim.

I think I see what you're saying a bit better now. I've often thought of the Fed as being passive regarding the monetary base, but had not fully considered that with regards to FFR. I think in practice you may be correct (which could fit with the expectations view), although in theory the Fed could garner more control.

A real test would be the Fed signaling one direction and acting the opposite or trying to move the rate opposite a market-determined direction. The latter seems more likely than the former but I doubt either will occur for some time.

Anyways, thanks for clarifying your view. It definitely gives me something to consider and research going forward.

Jazzbumpa said...

Joshua -

I agree. The two test cases are unlikely to ever play out in real life.

Thanks for being open to my ideas.

Cheers!
JzB

Unknown said...

JzB,

After considering your argument further, I'm persuaded that you're right about the Fed following the market in setting short-term interest rates (http://bubblesandbusts.blogspot.com/2012/08/markets-determine-interest-ratesuntil.html).

I still don't believe this undermines the case that the Fed could set rates at any level, if it chose to. The ECB potentially setting a ceiling on rates will provide some insight on that regard.

JW Mason said...

Hi Art.

I'm sorry I'm only just now seeing this post. This is actually the sort of conversation that needs to be happening, and I don't want to give the impression I'm not interested in your response here. It's a very valid & important point you raise.

Let me make two general responses.

First, in my opinion the Fed DID control long-term interest rates (or the availability of long-term credit, the more relevant but less easily measured concept) for a number of decades in the mid-20th century. We can say 1945-980 for simplicity. This was because of specific institutional and regulatory features of the financial system, which we can summarize as a tight link between bank reserves and total credit. Again, this link was the result of a specific regulatory and institutional environment. After the mid-1980s, this environment no longer existed.

Second, while I am a strong defender of the looking-at-picture methodology in historical macroeconomics (a defense of this vs. more formal time series approaches is one of the many hypothetical blogposts floating in my head) we do have to be cautious when the pattern we are looking at consists of only one or two trends. This is a useful thing you learn in an econometrics class -- take two series with unit roots (i.e. that follow a random walk) and run a regression of one on the other, and you can get an arbitrarily high r-squared even if there is no genuine correlation at all. Same with time trends. This doesn't invalidate the graph I presented since we have a good half-dozen distinct tightening and loosening episodes, but it is a problem when you are interested in longer-term trends. Many interest rates have fallen since 1980; is the Fed responsible? Hard to say.

Another point is that inflation has also fallen since 1980. I have some issues with calling x minus inflation "real" x, but setting those aside real interest rates really haven't fallen much, for long market rates, tho they have for the policy rate.

It doesn't really look as if market rates moved with the Fed rate, even before the Great Financialization. Actually, it looks like market rates are "sticky down".

Yes, that's exactly how I'd summarize it. So what do we disagree about? anything?

The Arthurian said...

Hi, JW. Good to see you.

Again, this link was the result of a specific regulatory and institutional environment. After the mid-1980s, this environment no longer existed.

Yeah, I didn't think about that when I was writing the post. And yet, the change in the regulatory and institutional environment was driven by the requirements of expanding finance, don't you think? I would have gone the other way, keeping and strengthening the environment, and hindering the expansion of finance. We should have been able to figure that out from the rise of interest rates (and the increase in debt) prior to the Volcker years. We should not have had to wait until the post-Volcker decline brought rates back to zero. We should not have had to wait for the crisis to teach us the lesson.

My measure of how good an economist is, is how much off-topic stuff I generate in response to him. You're good.


Second, while I am a strong defender of the looking-at-picture methodology in historical macroeconomics (a defense of this vs. more formal time series approaches is one of the many hypothetical blogposts floating in my head) we do have to be cautious when the pattern we are looking at consists of only one or two trends. This is a useful thing you learn in an econometrics class...

This doesn't invalidate the graph I presented since we have a good half-dozen distinct tightening and loosening episodes, but it is a problem when you are interested in longer-term trends.


"the looking-at-picture methodology"?

I'm a student of the economy. I look at graphs. (Granted, I have never taken an econometrics class.)

I think what you are saying here is that it is valid to look at a "short" period like your 25-year graph, because it shows a pattern that repeats several times...
and that it is less valid to look at a "long" period like my 1934-2012 graph because it shows a (larger) pattern that does not repeat... or seems not to repeat... or repeats on a scale that defeats data collection.

And yet, the larger pattern is built from the smaller, repeating pattern -- that, and on other factors. And to the extent that the one is built on the other, the smaller pattern *validates* the larger. I can't defend that statement, but you probably could.

Glad you liked my "sticky down" observation.