Normalizing the Fed Rate: This Time is Different
One hears that the Federal Reserve would like to normalize the Fed rate. What does that mean? How can we conceptualize that?
Well, there are many ways to conceptualize it, but here is a simple way using measures of the growth of nominal GDP (NGDP). In normalization, the Fed rate will rise to the growth rate of NGDP by the end of the business cycle.
First, we measure NGDP growth by multiplying M2 money stock by the velocity of the M2 money stock. Then take the annual percentage change of that multiplication. (link to FRED graph)
The other way to measure the growth in nominal GDP is by multiplying real GDP by the price level, then taking the annual percentage change. I will add this measure to the previous graph.
They are close.
Note that the current growth in NGDP is around 4% which is a level only seen around times of recession in the past.
Note the slow general decline since the 1980’s.
Normalization of Fed Rate
Now, how does the normalization of the Fed rate fit into this picture?
Let me add the effective Fed rate to the first graph, and make some comments.
Generally, the Fed rate will fall during a recession. Then after a recession, the Fed rate will climb below the growth rate of NGDP in order to heat up the economy toward full employment. Normalization is when the Fed rate rises up to or above the level at which NGDP has been tracking. This is a signal that full employment has been reached and that the desire to heat up the economy must stop. If the Fed rate goes above the NGDP growth rate, a recession is likely to form.
Note that before the crisis, NGDP was tracking around 7%. The Fed must have thought that they could raise the Fed rate safely to about 5%. But when the Fed rate reached 5%, NGDP growth fell to below 5% at the same time. The consequence may have triggered the formation of the recession.
Formula
What is the rate at which the Fed rate will normalize to NGDP?
Normalized Fed rate = natural real rate + inflation target
Natural real rate is related to the growth rate of real GDP. Inflation target is related to the current core inflation.
Another formula could be used.
Normalized Fed rate = % change in M2 stock + % change in M2 velocity
Better this formula…
Normalized Fed rate =yearly % change of (M2 stock * M2 velocity)
This Time is Different
Obviously the current business cycle shows a different pattern than seen in all previous business cycles. The Fed rate has fallen and can’t get up. Even though the Fed rate “would” normalize at a lower rate this time than seen since the 1950s, around 3%, it cannot even get started toward this low level.
NGDP is growing slow in spite of massive Federal Bank reserves. Money will just not accelerate through the economy… The weak momentum gives the Fed fragile reasons to raise the Fed rate.
Some causes that have contributed to the strangeness of this business cycle are… low labor share, higher debt levels, global disinflation, weak fiscal policy, lack of identifiable bubbles and weak domestic productive investment.
One might think that a recession is just not possible with the Fed rate so far below the NGDP growth rate.
Still, the economy currently is really very sick… in ways that we may not have seen before.
More …
The Fed is confused on where full employment is. One, full employment is not behaving like it used to. Two, full employment is lower than the Fed perceives from the CBO. The result will be that full employment sneaks up on the Fed. Recession will happen this time with the Fed rate much lower than the NGDP growth rate. These are interesting times.
Is it possible that lack of demand has reached a level that makes growth nearly impossible? Is this not the effect of the debt economy, Piketty’s gap of growth to asset income and tax policy? I realize conclusions are unprofessional, but you have raised the question why this time is different.
Is there some reason you chose to derive NGDP rather than simply choosing it in FRED ?
https://research.stlouisfed.org/fred2/graph/?g=21EB
Never mind. I suppose you did it to better illustrate your “formula” points.
James Sherry,
There is a lack of demand which comes from low labor share and higher debt both through this business cycle. I relate labor share to effective demand limit. But high debt has its suppression mechanisms on the momentum of the economy.
Marko,
Yes… the point was to look under the hood a bit and incorporate money stock and its velocity mostly. The M2 velocity is still falling year over year. This tends to depress interest rates. From the way I look at velocity. A 1% decline over one year would depress interest rates by 1.5% roughly. The falling velocity is another reason interest rates are low.
What cause, if any, has the payment of interest on reserve deposits had on the velocity of money?(Or, are reserve deposits not a part of M2?) What caused the sharp increase since 2010, as shown in your graph? And how much of that M2 money is outside the US economy?
The Fed looks at inflation as one of the factors when deciding whether to raise interest rates.
But currently, labor has almost no power to negotiate for higher wages. And the new jobs being created are generally of the low wage variety. Thus labor does not have the money to drive prices up.
We are In an environment where we are more likely to see deflated prices for discretionary purchases.
We used to consume the product of each others’s labor, now too much of what we consume comes from southeast Asia. Our domestic circular flows have been disrupted by Global Free Trade.
That disruption could have been predicted but economists assumed that demand was ever present. That the Fed could ALWAYS induce spending by lowering interest rates. None of them foresaw a time when substantial numbers of labor/consumers would max out their ability to borrow. There was a FLAW in the economists’ logic.
We are on the verge of the next recession. The Fed can not lower interest rates. Again they will use Quantitative Easing to try to stimulate the economy, but the largest effect of that technique was to make more money available for the speculators in stocks, commodities, and probably real estate.
The Fed should have normalized interest rates last year or the year before that. By this December we will have had 7 years of the Fed’s zero interest rates.
The ‘powers that be’ have created a first class disaster in the American economy. NOW they are afraid to act, where was that timidity before?
EL – A little help please? You have the formula for the “Normalized Fed rate” as:
Normalized Fed rate = % change in M2 stock + % change in M2 velocity
Over the past year M2 has grown at a 6.5% rate. While Velocity fell by 2%. So I get 4.5% as the “correct” rate.
Over the past quarter M2 has grown by 9% while Velocity has increased. The V rate of change for the Q was 1% (annualized). This would get the correct Fed rate close to 10%.
What am I missing??
bkrasting— I know you are asking Edward, but I put in my two cents worth.
You should only use the annual data as the monthly/quarterly data is too noisy. Monitor the short run data for signs that the trend of the long term data is changing.
Since Ed is using quarterly data, so there is a good lag. I use the monthly personal income data where there is a great relationship between that growth rate and yields.
Since I’m making comments, I use zero maturity money( MZ) rather than M1 OR M2 and the correlation between MZ velocity –based on monthly personal income– and bond yields is even better than M2 velocity.
Speaking of interest rates.. The CBO just came out with its latest forecast for % rates (short and long-term).
Who knows what the future will bring? I wish I did, I’d make a bundle if I could. The CBO forecasts track with that of SSA. So I guess these should be taken seriously.
Me? I think CBO has it wrong. There is no way that we will see the rate increases that CBO believes are in our immediate future. 4+% ten-year and 3+% short-term rates are not going to happen between now and 2020. We shall see.
The last slide of the report provides a forecast for the yield on SSTF assets. In July of this year SSA invested $186B for fifteen years at a lousy rate of 2%. The CBO thinks this yield will jump to 4.5%…….
CBO link:
https://www.cbo.gov/sites/default/files/114th-congress-2015-2016/presentation/50866-interestrates.pdf
JimH: “We used to consume the product of each others’s labor, now too much of what we consume comes from southeast Asia. Our domestic circular flows have been disrupted by Global Free Trade.”
I agree with the premise, but not the conclusion.
The problem is not free trade, but our trade deficit. We consume more than we produce. As such, we must import the excess. Worse, we have to borrow money (credit cards, car loans, mortgages, etc.) to pay for that excess.
It is not the global free trade that is the problem, but the ease of getting into debt and our willingness to get into debt.
One more note on % rates. Today Treasury sold three month Bills at a yield of 0.00%.
When has that happened before? Answer = NEVER!
There’s a hell of a lot of cash money sitting around right now.
The T-Bill results:
https://www.treasurydirect.gov/instit/annceresult/annceresult_query.htm?cusip=912796FP9
bk:
Time to start ye olde sub-prime Alt-A market up again. Tranche them, rate them, and insure them with CDS with a counter of a naked CDS. 2008 here we come . . .
The last BLS employment report was a mixed bag, at best. The Establishment Survey reports 142k more jobs (seasonally adjusted), but 24k were government jobs (generally a drain on the system, not added production), and we lost 13,000 manufacturing jobs.
The Household Survey, which I trust a bit more, shows 236k fewer people employed — 136k part-time and 100k full-time.
The civilian population increases 229k and the workforce lost 350k?
Then, you dig further into the numbers, and see that ALL of the job creation in the Establishment Survey was from the seasonal adjustment. There were actually 525k FEWER private jobs in September, not 118k more.
That’s “double-plus ungood.”
BKRasting,
You are right. I should have stuck with the formula in the graph…
Fed rate = yearly % change of (M2 stock * M2 velocity)
The formula that I put is not as good.
I will change it.
E.L. Tks for the clarification. I’m still confused.
Over the past year M2 is up 6.7% while V is falling at 2%. Multiply those two any you get -0.13%.
So the formula used says that rates should still be pegged at zero – Yes?
What if we had strong growth in M2 and V was increasing? Use +10% M2 and +5% velocity. Then even in these extreme conditions the rate would be 0.5%. To get to a 3% rate M2 would be growing at 30% and V would be growing by 10%. What happens when M2 rises by 20% but V remains constant? The answer is that the % rate is (always) zero.
I remain unconvinced that these formulas have much value.
Perhaps you are using the wrong numbers.
FedRate = (1 + yearly change of M2 stock) * (1 + M2 velocity) – 1
So your “M2 is up 6.7% while V is falling at 2%. Multiply those two any you get -0.13%” example becomes:
1.067 * 0.98 – 1 = 0.046 = 4.6%
More reasonable?
Warren – Yes it makes more sense when you add in those 1’s. But the result? The Fed rate should be 4.6% today? That can’t be right. We would be rapidly headed into a recession if we followed this formula.
What happens when we consider other scenarios for M2 and V? Assume that M2 is growing by 10% while V is climbing by 5%. What is the implied Fed rate? -99%???
I’m still missing something?
I think what you are missing is the implication that the Fed rate should have been going up for a long time now, and that since it has not, the economy has become addicted to the cheap money.
Remember that the premise of the post is that the FF rate will normalize to the NGDP growth rate , which is the growth rate of the product of money and money velocity , NOT the product of the growth rate of money times the growth rate of velocity.
Here’s the graph using M2 , along with potential NGDP for comparison :
https://research.stlouisfed.org/fred2/graph/?g=22VO
Marko – Thanks – I still don’t get it. You say it is the sum of M2 + V.
But the chart you give us is M2 * V.
So are we adding or multiplying?
Maybe this will help…
Start with M2a = 2, va = 2… The product is 4
Then raise M2 by 200%, and v drops by 50%,
M2b = 4, vb = 1. The product is still 4. So NGDP did not change. But if you add the percentages, 200% + -50% + 150%. This is wrong. NGDP did not go up 150%.
Then this formula works…
multiplier on change in NGDP = M2b/M2a * vb/va = 4/2 * 1/2 = 1
NGDP changed by a multiplier of 1, so it did not change.
The key is that a decrease in velocity will decrease the fed rate.
The big difference this time is government austerity. It’s like the 1930s again, at least as it was in Europe.
Lambert wrote:
“But when the Fed rate reached 5%, NGDP growth fell to below 5% at the same time. The consequence may have triggered the formation of the recession” (2006~).
I would not say the recession was caused/triggered) by FED´s rate hike. But FED was more late than usual behind the curve. The credit-binge and subprime ABS´s though took several blows due to rate hikes(liers loan&ARMs triggered the financial crises in the interbank-market(repo)). FED was sure the economy would continue to grow and Bernanke was in total denial about the housing-bubble. A bubble that started to leak 2005.
Had FED started rate-hikes earlier they would have been in better sync with the economy. The economy on the other hand would have showed it´s capacity to expand credit during a normal rate-hiking cycle.
Not raising rates earlier after 2000 was the biggest contributor to the deep recession(not only caused by the GFC).
A Case of Self-Induced Paralysis?* Ben S. Bernanke.
Bernanke about the japanese(BoJ) policy-errors during 10 years of deflation. Has he reasons to regret? He should!
http://www.princeton.edu/~pkrugman/bernanke_paralysis.pdf
Velocity is derived by calculating (money stock)/GDP
so
M2stock X M2velocity = GDP
so your fed rate formula just equals yearly % change GDP
Which is to say the Fed target follows the rate of
economic growth.
Jim,
When you say GDP, it is nominal GDP. And yes, Fed rate in a sense follows growth. First it lingers behind. Then catches up. But the Fed rate should not be higher than GDP growth unless core inflation is rising with momentum.
Why don’t you graph %change of NGDP instead of obscuring the fact that is what the blue curve on the graph is showing.
M2Velocity is derived from GDP and M2. It is not an independently measured quantity.
Jim – Correct. V is just a ratio of M2 and GDP. So when E says:
“Normalized Fed rate =yearly % change of (M2 stock * M2 velocity)”
He really means:
Fed rate = % change in NGDP.
Look at 2010. NGDP grew 4.5%, but unemployment was >9% for the year. A 4.5% Fed rate would have been a huge error. The formula produces garbage.
In addition, V can only be determined months after the current Q as the GDP data is unavailable to make a calculation. The Fed looks forward, not backward.
Possibly E.L. should draw a line through his second formula as well as the first.
BKrasting,
Of course it is % change in NGDP.
And the Fed rate should not equal that at all times. The fed rate has to stay low in order to heat the economy up to full employment. In the past, the fed rate would eventually rise to NGDP growth, but going over NGDP growth seems to assist in forming a recession.
But with the fed rate so far below NGDP growth now, is it even conceivable that a recession might occur?
EL – you now say:
“Of course it is % change in NGDP”
Of course? In the text you twice used V in formulas to make your case. In the comments you provided two more calculations using V. So why the “smoke and mirrors” in your formulas?
Now that we understand that your definition of “Normalized” is = to: “% change in NGDP”, can you show us what the “normalized rate” should have been the past six years using your formula? Then defend the results against the facts that during those 6 years unemployment was high, employment was low, wages were stagnant and inflation was very low.
BKrasting,
In the beginning of the post I say…
“here is a simple way using measures of the growth of nominal GDP (NGDP). In normalization, the Fed rate will rise to the growth rate of NGDP by the end of the business cycle.”
So of course these are measures of NGDP growth. There are no smoke and mirrors. I wanted to show that a declining velocity can lower a normalized Fed rate.
Now you are confused about what normalized means. The normalized Fed rate should not have been applied during the last 6 years, when the Fed rate was supposed to be below the normalized fed rate. Like I say in the post…
“Generally, the Fed rate will fall during a recession. Then after a recession, the Fed rate will climb below the growth rate of NGDP in order to heat up the economy toward full employment. Normalization is when the Fed rate rises up to or above the level at which NGDP has been tracking.”
So it is proper that the Fed rate has been below its normalized level. Eventually the Fed wants the Fed rate to rise to the normalized level of NGDP growth.