Taylor Rule Estimate: Fed Fund Rate Differential at a Whopping 6.8%

When Zero Hedge initially looked at the Taylor Rule Estimate for the Federal Fund Rate back in January, the prevailing consensus was that, even then, the Taylor-to-Fed Fund Differential was a whopping 6%. For those who wish to familiarize themselves with the Taylor Formulation, we suggest reading our initial thoughts, but in a nutshell Taylor's thesis is that the FOMC sets it fund rate target according to this rule:

i=r*+p+0.5(p-p*)+0.5(y-y*)

where i is the nominal fed funds rate, r* is the equilibrium real funds rate, p is the inflation rate, p* is the Fed's desired inflation rate, y is the (log) level of real GDP, and y* is the (log) level of real potential GDP. (Various banks introduce coefficients that change the weighing of some of the factors, but in its purest, this is what the original Taylor formulation reads).The default calculation assumes a target inflation of 2% and unemployment rate of 5%.

Now of course only Sweden seems capable to pull of a negative Fund rate, which is why monetary policy is futile these days, and the only alternative the Fed has is to promote liquidity via Quantitative Easing.

So the question now is whether after several hundred billion of agencies, MBS and treasuries has already purchased, has the Taylor Rule estimate improved at least marginally, or, in other words, has the Fed Fund Rate to Taylor spread tightened.

The answer is an emphatic no, implying that in as much as QE was supposed to be the last bastion of monetary policy, it has failed, and all it has achieved has been to maintain interest rates on mortgages and bonds within a reasonable bracket. And as long as the money multiplier remains suppressed due to the population's perception of the true economy (which is surprisingly objective, despite the media's 24/7 propaganda that everything is now better), this negative metric will persist, and eventually force Bernanke to acknowledge that his monetary policy has been a failure.

The first chart demonstrates a time-study of the Fed Funds Rate superimposed on top of the Taylor Rule Estimate: According to Taylor if monetary policy had free reign, the Fed would need to set rates at -6.55% right now in order to stimulate inflation.

The 6.8% spread between the Taylor and the Fund Rate is the widest it has been in over 2 decades, while the actual Taylor reading is now the lowest it has been since data has been accumulated in the early 1980s.

So now that QE is gradually being unwound and various liquidity measures are starting the be reeled in by the Fed, the key question becomes just what weapon does the Fed have left in its arsenal to stimulate the monetary base in the hope of offsetting prevailing deflation and unemployment?

Alas, hope and green shoots do not work here.

Comments

[Joseph L. S...:]

The Fed has now boxed itself into playing the role of Blanche DuBois in "Streetcar Named Desire": it must now depend on the kindness of strangers not to upset this delicately-balanced applecart...

Aug 19 10:56 AM

[JPSmith:]

Assume you meant "emphatic"....or does one need to be highly attuned to the feelings of others to understand Fed policy? :-)

"The answer is an empathic no"

Aug 19 11:56 AM

[WhoisJohnGalt:]

With all respect to Keynes, but looks like the best solution right now is to bite the bullet and delever on all scales: consumer, corporate, and eventually government. We have kid ourselves for decades with an ever-growing standard of life that was actually propped up by a house of credit cards. The savings rate will have to shoot up to double digits for a while if we want to live within our means. So if anything, squeaky clean lending standards should be enforced. It's not sexy profit for banks, but at least it's sustainable.

Aug 19 11:57 AM

[djackson:]

Nice piece, but tremendously sensitive to assumptions. Change inflation to core PCE and suddenly the gap drops to less than 1%. Using core is more in line with most economic thinking although I grant a good case can be made for CPI in certain environments. Another would be to agree with the El Erians of the world that we probably have a different employment outlook going forward than in the past decade. Change NAIRU to 6%, using core PCE, and suddenly the gap approaches zero. And what if the OKUN factor of 2 is too agressive? Change it to 1.5 and suddenly the TR says the fed funds rate should be 70 bps higher than it is. I am not trying to quibble with you, just pointing out that accepting default assumptions on your bberg (your default was CPI, wasn't it?) can lead to solutions most would regard as silly. Your body of articles suggests to me that you can do better than that.

Aug 19 02:49 PM

[Dialectical...:]

Let's just assume unemployment of 20% and voila... problem solved. No more deflation... My point is of course you can plug in new values to get new results. But if you take the same assumptions or variables and plot them with historical figures you STILL get a trend line that goes in the wrong direction. If 5% unempoyment was assumed in the past, you should use it now for apples to apples comparison. Anything else is hocus pocus.


On Aug 19 02:49 PM djackson wrote:

> Nice piece, but tremendously sensitive to assumptions.

[...]
> Change NAIRU to 6%, using core PCE, and suddenly the gap approaches
> zero.

Aug 19 05:40 PM

[Alok Swain:]

a) Good article, especially about the "Taylor Rule Estimate" about what the interest rate should be, if interest rates were to be allowed to go below zero.

b) It takes the "unemployment numbers" and "desired inflation" into account in making an estimate of interest rates.

c) Here the gap of 6.8% differential between the "Taylor Estimate" and fed funds rate is the amount of deflation that we are experiencing, that remains un-bridged.

Aug 19 08:49 PM

[Moon Kil Woong:]

The Fed's level of incompetence is galling and only comparable to the politician's incompetence at limiting deficit spending to a rational percentage of GDP (ever hoping for no deficit spending is an irrational expectation for these clowns).

I remember when a prominent European said we will not go to 0 interest rates and not engage in QE it's a path to hell everyone complained. Now Europe has tightened its belt, lowered its trade deficit and is coming out of a recession.

The US is still hopelessly embroiled in false government stimulus that continues to encourage trade deficits, consumption beyond our means, and federal deficits of monumental proportions. It seems that for the first time the US outdid Europe in socializing the economy. Perhaps Europe is learning the lesson that the US has just begun to learn. Real growth means a real economy.

As people all point out, "Sometimes people have to learn for themselves, or anotherwords the hard way.". The US may have to learn the very hard way what a government led economy results in: stagnation, high unemployment, inflation, massive deficits, falling standards of living, and an uncompetitive economy.

Aug 19 10:13 PM

[Fabien Hug:]

This is a model and we have seen during the past years how models work.
Some comment about the level of unemployment used in this model, but wasn't the computation of this index changed in the 90's?
What's important is that heavy decision makers are looking at them, you just have to use the same assumptions/data as they do, if you can.

Aug 20 10:08 AM

[djackson:]

The logic fails. Plugging in a new NAIRU assumption does not "solve" a deflation problem; it merely indicates in this case the potentially appropriate policy given the forecasted structural environment of employment. Further, if all you are interested in is projecting trendlines, of course just continue to use whatever assumptions you historically used. Of course, this ignores the fact that economic conditions do not remain constant and therefore provides no insight into appropriate policy going forward. By that logic, you should presumably be using the same OKUN data that was deemed relevant to the 1970s.


On Aug 19 05:40 PM Dialectical Materialist wrote:

> Let's just assume unemployment of 20% and voila... problem solved.
> No more deflation... My point is of course you can plug in new values
> to get new results. But if you take the same assumptions or variables
> and plot them with historical figures you STILL get a trend line
> that goes in the wrong direction. If 5% unempoyment was assumed in
> the past, you should use it now for apples to apples comparison.
> Anything else is hocus pocus.

Aug 20 10:12 AM

[Philman:]

The idea that we are going to face deflation is ridiculous. Inflation is already very evident so long as you avoid the government statistics which are massaged to show otherwise. Using the more honest 1980's inflation formula, shadowstats.com calculates the current inflation rate to be a little less than 6%, down from over 12% last year. That is reality.

Oh, but you're not including housing, you might claim. So what? Housing wasn't included on the upside, either. If we did include housing, we would have had a 20%+ inflation rate for the last 9 years. It was the Fed's policy not to mess with bubbles. Now, it is their policy to reflate them?

Taking rates down below zero by quantitative easing is punishment for the careful honest savers, and a free gift to the prolifigate spenders and wasters, who caused the worldwide economic Crisis. It shouldn't be done. It MUST NOT BE DONE! If it continues to be done, a point of no return will come, like in Weimar Germany 1919-23. I'm sure that if you run your ridiculous "Taylor formula" you would have claimed that the Germans, back in that era needed to print money, also. Well, they did, and, a few years later, people had lost so much faith in the German mark had devalued to 1 trillion to 1 compared to the pre-WW I value of the paper mark.

Aug 20 11:12 AM

[Dialectical...:]

You're right. The logic was specious, but it was actually meant to be facetious. The point is two of the variables are DESIRED inflation and DESIRED unemployment. If you want to find the "appropriate policy" for a "desired unemployment" of 20%, for example, you can. But my point is that you can't twiddle with the "desired" targets just to come out with a good number when those targets aren't anything close to what has historically been deemed appropriate. The formula itself may not function that well in the best of circumstances, but certainly it can't function at the outer ends of the scale, because all kinds of economic activity would change with (for example) 10% inflation and 20% unemployment. These changes would break the formula. I think it is instructive to note that the interest rate "should be" negative using normal assumptions from normal periods of economic activity. The fact that we are approaching a new normal is true, but not necessarily relevant to the point of the article.



On Aug 20 10:12 AM djackson wrote:

> The logic fails. Plugging in a new NAIRU assumption does not "solve"
> a deflation problem; it merely indicates in this case the potentially
> appropriate policy given the forecasted structural environment of
> employment. Further, if all you are interested in is projecting trendlines,
> of course just continue to use whatever assumptions you historically
> used. Of course, this ignores the fact that economic conditions do
> not remain constant and therefore provides no insight into appropriate
> policy going forward. By that logic, you should presumably be using
> the same OKUN data that was deemed relevant to the 1970s.

Aug 20 02:59 PM

[User 474650:]

Your assumption of a constant NAIRU is improbable. In recessions it is highly likely that factors in the labour market, such as hysteresis, and in the wider economy, such as a withdrawal of credit, can lead to a permanent increase in the natural rate of unemployment.

Your argument's assumptions imply a purely demand driven economy by keeping the NAIRU constant. When clearly there have been supply side drivers and effects.

Im not saying that tweaking the NAIRU until you achieve your desired results is the right way to go. I am merely highlighting that your very strong results hinge upon their assumptions massively and so are not so strong.

There's a reason why they let you change the NAIRU in that particular bloomberg page.

You also have failed to consider any adjustment for policy inertia. If you increase Rho, the split begins to fall considerably. As the real interest rates become negative its reasonably obvious that policy inertia will increase as central bankers become more "dovish" and they have less potential to decrease nominal interest rates on top of this.

On Aug 20 02:59 PM Dialectical Materialist wrote:

> You're right. The logic was specious, but it was actually meant to
> be facetious. The point is two of the variables are DESIRED inflation
> and DESIRED unemployment. If you want to find the "appropriate policy"
> for a "desired unemployment" of 20%, for example, you can. But my
> point is that you can't twiddle with the "desired" targets just to
> come out with a good number when those targets aren't anything close
> to what has historically been deemed appropriate. The formula itself
> may not function that well in the best of circumstances, but certainly
> it can't function at the outer ends of the scale, because all kinds
> of economic activity would change with (for example) 10% inflation
> and 20% unemployment. These changes would break the formula. I think
> it is instructive to note that the interest rate "should be" negative
> using normal assumptions from normal periods of economic activity.
> The fact that we are approaching a new normal is true, but not necessarily
> relevant to the point of the article.
>

Aug 21 07:28 AM