Monday, August 24, 2015

Forget the Sept. '15 Fed Tightening -- There Has Been a Stealth Tightening Since Sept. '14

August 24, 2015

Forget the Sept. ’15 Fed Tightening – There Has Been a Stealth Tightening Since Sept. ’14!

The late, great Milton Friedman used to preach that you don’t assess the degree of monetary policy restriction or accommodation by the level and movement of interest rates but rather by the behavior of monetary quantities. For Friedman, the monetary quantity by which he judged the stance of monetary policy was some definition of the money supply – currency and some of the deposit liabilities of the banking system. Meaning no disrespect to the greatest monetary theorist of the 20th century, but the monetary quantity I prefer to use to measure the stance of monetary policy is on the asset side of the banking system’s balance sheet -- the sum of banks’ cash reserves provided by the central bank and the loans and securities granted by the banking system. By my measure, the Fed has been gradually tightening its monetary policy at least since September 2014. Moreover, in recent months, Fed monetary policy has become downright restrictive.

This is illustrated in Chart 1. Plotted in Chart 1 is the behavior of the sum of commercial bank credit – loans and securities held by U.S. commercial banks -- and the cash assets held by these commercial banks – largely banks’ cash reserves created by the Fed. Past readers of my commentaries will recognize this monetary quantity as my concept (really Austrian economists’ concept) of “thin-air credit”. I refer to it as thin-air credit because cash reserves provided by the Fed are created, figuratively, out of thin air and so, too, is commercial bank credit in the fractional-reserve banking system that we have in the U.S. The red bars in Chart 1 represent the month vs. year-ago month percent changes in this monetary quantity. The blue line represents the three-month annualized percent changes in this monetary quantity. To put things in an historical context, the median percent change in the annual averages of this monetary quantity from 1975 through 2015 was 6.62%.
Chart 1
The year-over-year percent change in the sum of commercial bank credit and cash assets peaked in July 2014 at 9.74% (the red bar just to the left of the solid black vertical line in Chart 1). The Fed’s third round of quantitative easing (QE), which had begun in September 2012 and was terminated in October 2014, was still being phased down in July 2014. The year-over-year percent change in this measure of thin-air credit has trended lower since July 2014, slowing to only 4.45% as of July 2015. The more variable three-month annualized growth in this measure of thin-air credit also has been trending lower since July 2014, slowing to 1.25% as of July 2015. Against the backdrop of a 40-year median growth rate of 6.62% for this measure of thin-air credit, recent months’ growth rates in it are relatively low, implying that the Fed’s monetary policy has gone from  accommodative in July 2014 to restrictive in July 2015.

Plotted in Chart 2 are the year-over-year percent changes in the two components of this measure of thin-air credit – the cash assets held by commercial banks, primarily cash reserves provided by the Fed, and loans and securities held by commercial banks. This chart shows that key driver of the slowdown in thin-air credit growth since July 2014 has been the slowing in cash assets held by commercial banks, i.e., a slowdown in the Fed’s provision of bank reserves. In each of the three months ended July 2015, commercial bank cash assets have contracted vs. year ago. Although commercial bank credit growth has slowed in recent months, it nevertheless remained relatively robust in July 2015 at nearly 7% on a year-over-year basis.
Chart 2



It stands to reason that growth in bank reserves would slow as the Fed tapered and then ended its purchases of securities, the largest source of reserve creation. But ending Fed securities purchases would not necessarily result in a contraction in bank reserves. But as Chart 3 shows, this is exactly what occurred. While the year-over-year dollar change in Fed securities holdings has narrowed, the year-over-year dollar change in bank reserves has actually contracted in six of the past nine months.
Chart 3


Chart 4 shows that as the year-over-year dollar changes in in the total supply of reserves, Fed security holdings plus other factors such as Federal Reserve float, have become smaller since the Fed began phasing out QE3, the year-over-year dollar changes in factors other than bank reserves absorbing the supply of bank reserves have increased. Some of these absorbing factors such as currency demanded by the public and Treasury deposits held at the Fed are beyond the Fed’s control. Back in the pre-QE era, the Fed used to engage in what were termed “defensive” open market operations to offset undesired absorption of reserves from these uncontrollable factors.







Chart 4

But as shown in Chart 5, there have been some factors absorbing the supply of reserves under the Fed’s control that have been increasing. These two reserve-absorbing factors controlled by the Fed are reverse repurchase agreements with government securities dealers and money funds and term deposits offered to commercial banks and other depository institutions. For reasons known only to it, the Fed began “experimenting” with these reserve-absorbing operations early in 2014 as it began tapering its reserve-supplying securities purchases. The Fed then stepped up its reserve-absorbing operations from October 2014 through February 2015, after it had ceased its QE securities purchases. Go figure.
Chart 5
The upshot of all this is that from a monetary quantity perspective, monetary policy has tightened significantly from a year ago. Growth in the credit being created by the private banking system has returned to near its long-run median rate after having been severely constrained during and after the 2008 financial crisis. But total thin-air credit growth is below its long-run median rate because of the slow growth or contraction in the Fed’s contribution. In my opinion, the Fed was correct in phasing down its securities purchases when it commenced to do so. Growth in total thin-air credit was excessive at that time. But without the Fed having any clear understanding of how many securities to purchase when it commenced its three QE programs, it had no idea of by how much it should taper its securities purchases. The Fed’s management of securities purchases and reserves growth needs to be guided by what was happening to growth in total thin-air credit – banks’ contribution and the Fed’s. There is absolutely no evidence that the Fed has been guided by this. As a result, monetary policy has gone from one extreme – overly accommodative – to the other – overly restrictive. The more things change, the more they stay the same.

Now, let’s talk about something important – the recent swoon in the U.S. stock market.  Plotted in Chart 6 are the year-over-year percent changes in monthly observations of the Wilshire 5000 stock market along with year-over-year percent changes in monthly observations of thin-air credit. Although growth in the Wilshire 5000 peaked about six months before thin-air credit growth peaked, the peak in the Wilshire 5000 growth occurred around the time the Fed began tapering its securities purchases. To the degree that financial markets anticipate events, it is conceivable that the stock market movers and shakers anticipated that the Fed’s phasing out of its securities purchases would eventually have an adverse effect on stock market values.
Chart 6
After massive purchases of securities by the Fed, why might the cessation of these purchases have an adverse effect on stock market values? When the Fed purchases securities, the sellers are government securities dealers. Dealers do not hold the quantities of securities that the Fed was purchasing during QE, especially after the 2008 financial crisis. In order to accommodate the Fed’s demand for securities, dealers bid for securities that their customers held. Dealers’ customers are large institutions, e.g., pension funds, insurance companies, mutual funds, and banks. The dealer community was an intermediary between the Fed and large financial institutions. So, when the Fed purchased securities, ultimately, large financial institutions sold securities and received cash. It is unlikely that these large financial institutions were to content to give up an earning asset for cash that has no explicit nominal return. Rather, these large financial institutions used the cash it received indirectly from the Fed, cash created figuratively out of thin air, to purchase other earning financial assets. Some of these assets purchased by large financial assets might have been equities. Some of these assets might have been bonds issued by corporations for the express purpose of financing their share buybacks. The Fed’s creation of thin-air credit likely financed, directly or indirectly, increased purchases of riskier financial assets, which, in turn, boosted the prices of these riskier assets. If so, then the Fed’s cessation of securities purchases and contraction in its contribution to thin-air credit would reduce the demand for riskier assets, which would have an adverse effect on their prices. So, why did I wait until now, after the recent swoon in the stock market, to tell you that the Fed’s cessation of QE would have an adverse effect on the stock market? Well, truth be told, I did alert you to this probability back on November 17, 2014, in a commentary entitled “2015 Is Shaping Up to Be a ‘Turkey’ of a Year for the U.S. Economy and Stock Market”.

But the truth also be told, I had no idea that a stock market swoon would occur in August 2015. In fact, I had no idea a stock market swoon would occur at any time in 2015. But it did appear to me back in November 2014 that thin-air credit growth was likely to decelerate sharply in 2015 compared to 2014. And that a sharp deceleration in thin-air credit growth would have adverse effects on nominal aggregate demand and the value of risk assets. The adverse effect on the value of risk assets appears to be upon us. Of course, the collapse of the Chinese stock market has played a large role in the decline in prices of U.S. risk assets, perhaps a larger role than the sharp deceleration in U.S. thin-air credit. Weather-adjusted, U.S. aggregate demand has held up relatively well. I suspect that will change in the fourth quarter of this year.

In the face of very low goods/services price inflation and weak wage growth, the recent U.S. stock market rout is likely to postpone the previously-anticipated September 2015 Fed rate hike. It is too early to expect QE4. But the Fed certainly has the leeway to restart securities purchases if need be. In other words, take some comfort in a Yellen put.

Paul L. Kasriel
Econtrarian, LLC
Senior Economic and Investment Advisor
Legacy Private Trust Co. of Neenah, WI
1-920-818-0236



Tuesday, July 14, 2015

If the Fed Raises Policy Interest Rates in 2015, It's Likely to Be One and Done

July 14, 2015

If the Fed Raises Policy Interest Rates in 2015, It’s Likely to Be One and Done

Last week, Fed Chairperson Yellen indicated that the Fed was likely to raise its policy interest rates sometime before year end. Given the behavior of the sum of commercial bank credit and depository institution reserves at the Fed in the past three quarters, it is a mystery to me why the Fed would be contemplating a policy interest rate increase at this juncture. But if this is something the Fed just has to get out of its system, then the first increase is unlikely to be followed by a second interest rate increase for some time.

To briefly review my focus on the sum of commercial bank credit and reserves at the Fed , because depository institutions (commercial banks, savings institutions and credit unions) are required to hold only a fraction of the amount of their deposits as reserves at the Fed, they are able to create credit by an amount that is a multiple of their reserves. This credit is figuratively created out of thin air. The reserves that depository institutions hold at the Fed are, in fact, created by the Fed. When an asset item on the Fed’s balance sheet increases, say a Fed purchase of securities, a liability item on the Fed’s balance sheet also increases. One liability item on the Fed’s balance sheet is reserves owed to depository institutions. So, these reserves, which serve as the “seed money” for the credit created by depository institutions, also are figuratively created out of thin air.

Entities typically borrow in order to purchase something – a good, a service, a financial instrument. When the credit that is financing purchases is created out of thin air, net spending in the economy increases as the borrowers increase their spending and no one else cuts back on his/her spending. In contrast, when the credit that is financing purchases emanates from new saving, then the change in net spending in the economy is zero. Saving is the act of curbing one’s current spending and transferring this purchasing power to a borrower. So, the borrower increases his/her current spending and the saver decreases his/her current spending.

Plotted in Chart 1 are the year-to-year percent changes in the sum of depository institution credit and reserves held at the Fed along with the year-to-year percent changes in nominal Gross Domestic Purchases. Gross Domestic Purchases are expenditures by U.S. households, businesses and government entities on newly-produced goods and services. So, expenditures on imported goods and services are included in Gross Domestic Purchases while expenditures by foreign entities on U.S.-produced exports of goods and services are excluded. From 1953 through 2014, the contemporaneous correlation between percent changes in thin-air credit and percent changes in Gross Domestic Purchases is 0.59 out of a maximum possible 1.00. Gross Domestic Purchases does not include the value of used products purchased (e.g., used cars and existing homes) and does not include the value of financial instruments purchased. If a measure existed that added these excluded items to Gross Domestic Purchases, it is likely that the correlation between its changes and changes in thin-air credit would be even higher than 0.59.









Chart 1

So, now you know why I am obsessed with thin-air credit with respect to monetary policy. The behavior of thin-air credit plays an important role in behavior of nominal spending in the economy. If the growth in thin-air credit is rapid, the risk of the economy overheating and/or the formation of asset-price bubbles would be heightened. If growth in thin-air credit were rapid, increases in Fed policy interest rates would be entirely appropriate in order to prevent a rise in the inflation rate of goods/services and/or asset prices.

But, as shown in Chart 2, growth in a subset of total thin-air credit, the sum of commercial bank credit and reserves at the Fed, has been decelerating in recent quarters and is below its long-run median growth rate. Chart 2 contains growth rates in the sum of commercial bank credit and reserves at the Fed. Commercial bank credit accounts for over 80% of total depository institution credit. The Fed reports commercial bank credit weekly with a one-week lag, whereas it reports total depository institution credit only quarterly with about a one-quarter lag. The long-run median growth rate in the sum of commercial bank credit and reserves at the Fed is 7.1%. As shown in Chart 2, growth in the sum of commercial bank credit and reserves at the Fed was 6.0% in Q2:2015 vs. Q2:2014. Moreover, the growth in the sum of commercial bank credit and reserves at the Fed has been trending decidedly slower starting in Q4:2014, when the Fed terminated its QE program. Quarter-to-quarter annualized growth in the sum of commercial bank credit and reserves at the Fed slipped to 5.1% in Q2:2015 and its trend growth has been decelerating since Q4:2013. This current below “normal” growth in the largest component of thin-air credit implies that the dangers of a near-term overheated economic environment and/or inflation of an asset-price bubble are low.


Chart 2

The principal factor accounting for the recent slowing in the growth of thin-air credit is the Fed’s contribution, reserves at the Fed. This is shown in Chart 3, in which 3-month point-to-point annualized growth rates are plotted instead of growth rates of quarterly averages. Starting in November 2014, the month after the Fed’s QE program ended, the three-month annualized percent changes in reserves at the Fed, with two exceptions, have been in negative territory. That is, since the end of QE, reserves at the Fed have contracted on net. But the bank-credit component of thin-air credit also has demonstrated a slowing growth trend. After surging in the three months ended January 2015 at an annualized growth rate of 10%, growth in commercial bank credit slowed to a below “normal’ rate of 6.5% in the three months ended June 2015.







Chart 3

Again, the current behavior of thin-air credit does not portend an acceleration in the rate of price increases for goods/services or assets. And the recent behavior of these price changes would not seem to warrant concern by the Fed. Consumer price inflation, as represented by the Personal Consumption Expenditure price index, remains tame. In the three months ended May 2015, this measure of consumer price inflation had increased at an annualized rate of 2.2%, recovering from the price declines earlier in the year due to falling energy prices (see Chart 4). Commodity prices of all stripes have been trending lower in recent months, as shown in Chart 5. And asset prices are not racing to the moon (See Chart 6).













Chart 4
Chart 5

Chart 6
So, current inflationary pressures are quite mild here in the U.S. The current rate of growth in U.S. thin-air credit is below its “normal” rate, suggesting that credit creation is not fostering a future surge in U.S. inflation. And the global inflationary environment appears equally tranquil, if not more so. The Chinese economy, which already had experienced a growth slowdown, will now be negatively affected by its recent stock market swoon. And Europe is not exactly booming, Greece aside. Given all this, it is not clear what is motivating the Fed’s desire to raise its policy interest rates sometime later this year. Whatever the motivation, if the Fed does pull the interest-rate tightening trigger in 2015, it will not likely do so again for many months thereafter. In other words, for Fed interest rate hikes in 2015, it’s one and done.

Paul L. Kasriel
Senior Economic and Investment Adviser
Legacy Private Trust Company of Neenah, WI
Founder, Econtrarian, LLC
1-920-818-0236



Monday, July 6, 2015

The June Unemployment Rate Fell to 5.3%, but ...

July 6, 2015

The June Unemployment Rate Fell to 5.3%, but …

It seems as though there usually is a “but” when the unemployment rate falls by more than expected by the cognoscenti. The consensus estimate for the June unemployment rate was 5.4%, down from May’s 5.5%. Instead, the BLS reported that the June unemployment rate fell to 5.3%, its lowest reading since April 2008. Might this be cause for celebration if one were not an economic masochist? Au contraire, according to the mainstream media talking heads.  After reporting the larger-than-expected decline in the June unemployment rate, the talking heads quickly inserted the “but” – but the labor force participation rate fell three-tenths of a point to 62.6%, its lowest since October 1977. The simultaneous decline in the unemployment rate and the labor force participation rate is consistent with the notion that the unemployment fell because people dropped out of the labor force due to discouragement over poor job prospects. So, recork the champagne, right? Wrong.
Before getting too deep into this, let’s define some terms. The “headline” or U-3 unemployment rate is the ratio of the number of people who are unemployed to the civilian labor force. Of course these numbers are estimates “blown up” from responses solicited from a relatively small sample of households. To be considered unemployed, one has to have been available to work if the opportunity to do so had arisen and have actively sought employment. The labor force is defined as the sum of people 16 years old or older who are employed or are unemployed (unemployed, i.e., according to the definition given above).

Let’s run through a couple of numerical examples of how the unemployment rate might fall. Let’s assume that the total number of “officially” unemployed is 5 million and the labor force is 100 million. In this case the unemployment rate would be 5% (5/100 x 100). Now let’s assume that the number of unemployed falls by 1 million to a total of 4 million and the labor force increases by 1 million to a total of 101 million. In this case, the unemployment rate falls to 4%, rounded. Alternatively, suppose that the number of unemployed again declines by 1 million to a total of 4 million and the labor force also declines by 1 million to a total of 99 million. Lo and behold, the unemployment rate again falls to 4%, rounded. By the laws of arithmetic, if both the number of unemployed and the number of the civilian labor force decline, the unemployment rate also will decline if the percentage decline in the number of unemployed is greater than the percentage decline in the number of the civilian labor force. And, arithmetically speaking, this is what occurred in June. The number of unemployed fell by 375 thousand or by 4.3% as the civilian labor force fell by 432 thousand or by 0.3%, resulting in a decline in the unemployment rate of two-tenths of a percent.

The question arises as to whether the drop in the number of unemployed was related to the drop in the labor force. For example, did 375 thousand people who in May were classified as unemployed become discouraged with their job prospects and throw in the towel with respect to seeking employment. If so, these people would no longer be categorized as unemployed if they had stopped looking for a job. And if these 375 thousand people were no longer looking for a job, they would no longer be counted in the civilian labor force. Thus, the labor force number would drop, too. This decline in the June number of unemployed and the labor force being due to discouragement of job prospects is just an hypothesis. We cannot make this determination simply by observing the simultaneous declines in both the number of unemployed and the labor force.

When a decline in the unemployment rate is accompanied by a decline in both the number of unemployed and the labor force, media talking heads look to the participation rate for clarification as to why the unemployment rate fell. For the life of me, I do not know why they look to the participation rate for clarification, but they do. The labor force participation rate is defined as the number of people in the civilian labor force (i.e., the sum of the number of employed and unemployed) as a percent of the population 16 years old and older, or the “potential” labor force. All else the same, then, a decline in the labor force would result in a decline in the labor force participation rate. As I noted above, in June, the labor force participation rate declined by three-tenths of a percent to decades low of 62.6%. But this does not tell us why the labor force dipped in June. Perhaps people became discouraged with their job prospects and dropped out of the labor force. Perhaps not. Perhaps people dropped out of the labor force because they retired. Perhaps they went back to school. Perhaps they dropped out of the “official” labor force to perform the most important of jobs, rearing a newborn child. So, the behavior of the labor force participation rate does not provide us with much “color” regarding the behavior of the headline unemployment rate. A decline in the unemployment rate and a decline in the labor participation rate are necessary conditions for an explanation of the decline in the unemployment rate being related to people dropping out of the labor force because of discouragement due to poor job prospects. But they are not sufficient conditions. In other words, more information is required to make such a judgment.


The critical information regarding the sufficient conditions is contained in Table A-1, “Employment Status of the Civilian Population by Sex and Age”, of the Household Survey section of the Monthly Employment Situation. Line item 8 in this table is [number of people] “Not in labor force”. Line item 9 in this table is [number of] “Persons who currently want a job”. If the principal reason people are dropping out of the labor force is discouragement over job prospects, then we ought to see approximately equal increases in the number of people not in the labor force and the number of those people not in the labor force expressing an interest in obtaining employment. Chart 1 shows the month-to-month changes in the number of people not in the labor force and the number of these people desiring a job.
Chart 1
In June, the number of people not in the labor force increased by 640 thousand. Of that 640 thousand, there was only an increase of 18 thousand that desired a job now. This does not comport with the hypothesis that the decline in the June unemployment rate was the result of people dropping out of the labor force because of discouragement over job prospects.

But wait. There’s more. There’s Table A-15, “Alternative Measures of Labor Underutilization”, in the Household Survey section of the Monthly Employment Situation. One of the measures of labor underutilization is the U-5 unemployment rate which takes into consideration not only the “officially” unemployed but those unemployed categorized as “marginally attached” to the labor force.  Persons marginally attached to the labor force are those who currently are neither working nor looking for work but indicate that they want and are available for a job and have looked for work sometime in the past 12
months. Discouraged workers are included in this category of marginally attached workers. If the headline unemployment rate (the U-3 measure of labor underutilization) and the participation rate were falling primarily because people were dropping out of the labor force out of discouragement, then the U-5 measure of labor underutilization, which takes into consideration discouraged workers and others who are only marginally attached to the labor force, would not decline and might even increase.  Chart 2 shows the month-to-month changes in both the U-3 (headline) and U-5 measures of the unemployment rate. In June, both the U-3 and U-5 measures declined by two-tenths of a percentage point, suggesting that job discouragement did not account for the decline in the headline unemployment rate.
Chart 2




I don’t know why the media talking heads do not look at line items 8 & 9 in Table A-1 and at Table A-15 in the Monthly Employment Situation report. Perhaps it is because they look only at the BLS press release, which does not include these items. I place little weight on the Monthly Employment Situation in assessing the current state of the U.S. economy. The data are based on samples, not universes. The data get revised many times over. But the media talking heads ought not to compound errors by ignoring relevant information in the report bearing on their hypotheses.

Paul L. Kasriel
Senior Economic and Investment Adviser
Legacy Private Trust Co. of Neenah, WI
Econtrarian, LLC

1 920 818 0236