">
Home
Oct 7, 2017 Employment Explosion
June 26, 2017 Why's the FED Panicking?
May 25, 2017 LFPR anyone?
Apr 26, 2017 What's up with the FED?
March 10, 2017 Feb Employment Situation
Oct 10, 2016 Tax Burden
Aug 1, 2016 Here Comes the Debt
June 26, 2016 Moribund US Economy
June 16 2016 Labor Update
Mar 10, 2016 Spring Renewal for Labor Markets?
Feb 21, 2016 GDP Gap
Feb 16, 2016 FED and Monetary Policy
Jan 19, 2016 Employment Gap Age Groups LFPR
Jan 10, 2016 A look at the Employment Situation
Dec 30, 2015 Fed Funds Rate up 25 Basis points...so what?
Dec 15, 2015 Fed Funds on the rise? Has Yellen 'Fell-in'?
Oct 15, 2015 Labor Markets Seven years of misery
Oct 6, 2015 Sept: Horrible Month for Labor
Sept 30, 2015 The FED: Interest Rate Angst
Sept 11, 2015 FED on the Monetary Policy Front
July 31, 2015 Trade and Foreign Exchange Rates
July 20, 2015 Economic Growth?
July 10, 2015 Labor Picture by Age Group
July 2, 2015 Disastrous Month in Labor Rpt
June 19, 2015 Minimum Wage - Income Distribution
Jun 5, 2015 Encouraged Worker Effect
May 8, 2015 Updated Employment Situation for April
May 4, 2015 Languishing Labor Markets
Apr 7, 2015 LFPR Doldrums on the Labor Front
March 8, 2015 Less than Zero Interest Rates - Trade War
2014 Articles
2013 Articles
2012 Articles
Dec 31, 2012 Fiscal Cliff --- Increased Spending
Dec 24, 2012 Fiscal Cliff---Rising Revenues with current tax cuts?
Dec 13, 2012 November Jobs Report
Nov 28, 2012 Regulation and the Financial System
Nov 17, 2012 Employment Escarpment - Moving the Jobs Needle
Oct 31, 2012 Update on Shale Gas and Tight Oil
Oct 11, 2012 Restructuring of an Industry: US Light Vehicles
Sep 4, 2012 Resuscitating the Moribund US Economy
August 4, 2012 Unemployment Rises Again
July 21, 2012 Misguided Fiscal Policy: Is it a case of fool’s gold, or the Consequences of Economic Ignorance?
July 6, 2012 Let Freedom  Ring!!! The Shale Gale
June 26 Productivity Macro
June 11, 2012 Painted into Corner
June 4, 2012 Encouraged Worker Effect
May 28, 2012 European Honeymoon Over
May 14, 2012 Back to Basics
May 4, 2012 Labor Force Participation Rate Shrinking
Apr 26, 2012 Income Distribution
Apr 15, 2012 Energy Independence
Apr 6, 2012 Jobs Jobs Jobs
Mar 27, 2012 Gas Prices Killing Economic Growth
Mar 15, 2012 Rough Road or Smooth Sailing?
March 9, 2012 Employment Challenges Ahead
Mar 6, 2012 Stalled US Economy?
Mar 1, 2012 FED Profitabiility
Feb 22, 2012 Population Changes
Feb 13, 2012 Bernanke on Unemployment
Feb 8, 2012 Lower Unemployment - Bad News?
Feb 3, 2012 Chinese Miracle???
Jan 12, 2012 Low Interest Rates - Why so low?
Jan 9, 2012 Labor Force Participation Rate
2011 & 2010 Articles
Introduction
About us
Links of Interest
Straw Poll
Definitions & Miscellaneous
 

2012 Volume Issue 4


Economic Newsletter for the New Millennium

 

March 1, 2012

 

Editor
Donald R. Byrne, Ph.D.
dbyrne5628@aol.com

 

Associate Editor
Edward T. Derbin, MA, MBA
edtitan@aol.com


For a downloadable version, click here


FedProfitability3-1-12.pdf



...a bit more compressed version of same


FedProfitability3-1-12 compressed.pdf





WHY IS THE FEDERAL RESERVE SYSTEM (FED) SO PROFITABLE?


 …and the Consideration of Some of the Resulting Implications (Part One)

 

This is the first in a short series of articles on this website focusing on the profitability of the Federal Reserve System (FED) and a variety of questions and implications flowing from such profitable operations.

 

The Federal Reserve System (FED) has nearly always been highly profitable both in an absolute and relative sense.  This article will examine why it is so.  In the last few years those profits have risen significantly upward despite the fact that the FED was supposedly buying the “junk”, called toxic assets, in order to bail out a variety of financial institutions here and abroad.  Given the high rates of compensation with which the executives of these bailed out institutions were rewarded, they should have known better and not been so naïve or so greedy as to chase risk in order to seek higher yields.

 

Remember TARP, the QEs 1-3, Maiden Lanes 1-3, etc., etc.  The remaining articles in this series will be posted in the very near future.  Keep tuned in!!

 

Tracing the Fed’s profits and putting their size into perspective

 

What happens to those profits? The FED uses some of its profits to match the capital paid in required of member institutions resulting in a doubling of the size of total capital which is analogous to owner’s equity for private sector ‘for-profit’ firms.  The Capital Paid in is the total cost of stock the members acquire.  Remember that the stock of the FED does not trade in the financial markets.  The member institutions of the Federal Reserve System receive a 6% dividend on their Capital Paid in. 



After meeting all such requirements, the balance of the FED’s profits are transferred to the U.S. Treasury and are included under corporate profits taxes in the social accounts such as the National Income and Product Accounts (NIPA).  









Corporate Taxes Including FED.jpg


 

This transfer of the remaining profits to the U.S. Treasury is really a subsidy to the Treasury and in effect a hidden tax on the financial institutions that are members of the Federal Reserve System.  As we have seen in other articles on this website, if not taxed directly as in the case of the Personal Income Tax, the final burden of all taxes is born by households who ultimately bear the burden of all taxes whether directly or indirectly from the forward and backward shifting of taxes no matter where the initial incidence falls.

 

TAX PROPOSALS HERE, TAX PROPOSALS THERE, TAX PROPOSAL EVERYWHERE…BUT HOW DO WE ANALYZE THEM?

November 8, 2011
http://www.econnewsletter.com/100101.html  


"One of the best minds of this past century in the area of public finance, the now deceased Richard A. Musgrave, a former professor of mine many years ago, would lament in his lectures and in his text book, The Theory of Public Finance (1959), that no matter what the law says, who actually is the ultimate bearer of all or part of the profits tax, can differ dramatically from one case to another.  After going through a kind of verbal games theory of analysis, he concluded that one guess is probably as good as another and perhaps the assumption as to who on average bears the corporate profits tax would be one-third of the final burden resting on the buyers (higher prices), one-third on the equity capitalists (stockholders in the form of lower dividends), and one third on the other productive resources such as labor (reduced wages/compensation).  Imagine that, labor ultimately bearing one-third of a corporate profits tax on the firm employing labor."

 

The following chart shows the relative importance of this disposition of the FED’s remaining profits transferred to the Treasury.







FED portion of Corporate Taxes.jpg



The FED is indeed [historically] the most profitable firm in the nation as measured by its rate of return on equity (ROE) as the following chart shows.  Not even ExxonMobil with its cartelistic market power and a typical ROE in excess of 30% can come close to matching the profitability of the FED.







FED Return on Equity (ROE).jpg









FED Profits.jpg



LARGE PROFITS BECOME EVEN LARGER


In response to the Financial Fiasco of 2008, the FED loaded up with such things as CDOs (consolidated debt obligations) including MBSs (mortgage backed securities).  To help you see this transformation, you will find below several comparisons of the FED before the crisis of 2008 and since it intervened in the financial markets.  This transformation could have been seen step by step by watching the weekly releases of the by the FED such as the H.4.1

 

As the FED acquires assets, it has an expansive effect of the monetary base which in turn increases the capacity of the financial system to create more money and credit if the credit creating institutions in the system find it profitable to do so.

 

Most of such changes in the monetary base are found under the term ‘Reserve Bank Credit’ in the H 4 1 releases and includes:

 

Securities held outright (U.S. Treasury securities, Federal agency debt securities, and Mortgage-backed securities)

Repos (or Repurchase Agreements)

Loans (Primary, Secondary, Seasonal, Credit extended to AIG, Term Asset-backed Loan Facility (TALF)

 

Portfolio Holdings of Maiden Lane(s) 1-3, etc.








FED Reserve Bank Credit.jpg










FED Return on Assets (ROA).jpg










FED Balance Sheet H.4.1 Securities held outright.jpg








FED Balance Sheet H.4.1 Treasurys.jpg











FED Schedule of Maturities.jpg








FED Balance Sheet H.4.1 MBS (Mortgage Backed Securities).jpg










Liquidity Portfolio 2008.jpg









FED Balance Sheet H.4.1 Federal Agency Debt Securities.jpg




Having graphically viewed this transformation of the FED’s portfolio and related activities, let’s first see why the FED was profitable before its intervention of 2008, and since then as well due to the extensive intervention in the financial markets by the FED.  We will use the jargon of financial theory and analysis to achieve this.  Later issues in this series will focus on related issues.  
 

FINANCIAL THEORY AND ANALYSIS


The reader is encouraged to re-read other relevant issues of this newsletter on this or its archival websites, especially those listed at the end of this article.


Operating and financial leverage

Operating leverage is defined as the ratio of profits to assets.  That ratio is referred to as Return on Assets or ROA.
 
The degree of operating leverage or ROA ranges over a wide spectrum for firms.  While some financially successful firms may experience a relatively high 10% annual return on assets, others experience a ROA of less than 1%.  There are many factors that determine the profitability of a firm and its ROA, but one extremely important factor is the degree of competition they face in the markets in which they operate, both product or goods and services markets and productive resource markets such as labor and financial markets. 
 
As a general rule, the lower the degree of operating leverage a firm experiences, the higher the degree of financial or debt leverage it will employ.  The reason for this is that the two degrees of leverage, operating and financial, determine the firm’s Return on Equity or ROE (the rate of return to the stockholders on their equity investment in the firm) for any given level of profits.

In much of manufacturing, successful firms will experience a much higher degree of operating leverage or ROA such as 6% or 8% than in much of the financial services industry, especially in the depository institution portion of the financial services industry, where a ROA is often less than 1%.  If no debt (liabilities) were employed, in other words, zero financial or debt leverage were used, the Return of Equity (ROE) would be the same as the Return on Assets (ROA). 

To earn reasonable rates of return to equity investors, a firm with a low degree of operating leverage  (low ROA or low ratio of profits to assets) would have to employ a high degree of financial leverage (high ratio if liabilities to assets).  Sure enough, depository financial institutions who create most of the money and a significant amount 0f the credit in our financial system, are considered successful if their return of assets is 1%.  This would be dismal for a manufacturing firm which employs a much lower degree of financial leverage.  But the typical depository financial institution uses a very high degree of debt or financial leverage often 90% or more (i.e. liabilities equal 90% or more of assets which means that owners equity is often less than 10% of assets). 

Of course with such a thin buffer against bankruptcy, WELL MANAGED AND PRUDENT DEPOSITORIES SUCH AS COMMERCIAL BANKS CANNOT AND SHOULD NOT BEAR MUCH RISK!  The ongoing financial crisis laid bare the ineptness, of highly compensated executive management of these institutions that the FED decided it had to “bail them out.”  In our example, a ROA of 1% could be leveraged up to a ROE of 12.5% if a very high degree of financial leverage were employed.  In the example just given, liabilities being 92% of assets (owners’ equity is only 8% of assets), a relatively low ROA or operating leverage of 1% results in a reasonable ROE even if profits are only 1% of assets.  BUT THE DEPOSITORY MUST FORGO ANY SERIOUS AMOUNT OF RISK BEARING AND NOT CHASE YIELDS, OR EXECUTIVE MANAGEMENT IS FLIRTING WITH BANKRUPTCY! 

The FED has a much higher [rate of] operating leverage for a depository financial institution.  Granted it is not a typical depository.  The typical well run depository financial institution has a return on assets or ROA (measure of operating leverage) of around 1%.  In order to earn a reasonable return on owners’ equity or ROE, depositories such as commercial banks employ a relatively high degree of financial (or debt) leverage (as explained immediately above)..  This means that liabilities (for the most part) are typically about 90-92% of assets, with owners’ equity being about 8-10% of assets. 

Remember the balance sheet identity that the sum of assets must equal the sum of liabilities plus owners’ equity.  

A = L + OE

With financial leverage reflected in liabilities at 90% of assets, AN ROA OF 1% is leveraged by a multiple of 10 and yields a 10% ROE. 







Financial Institutions Return on Assets (ROA).jpg










Financial Institutions Return on Equity (ROE).jpg




As the chart above shows, the FED experiences a much higher ROA than do private sector depositories, such as commercial banks.  In 2006, a pre-crisis year through the just ended 2011, the FED’s ROA ranged from about 1.5% to nearly 3.5% as compared to a typically successful commercial bank’s 1% ROA.  The FED employs a much greater degree of financial leverage as well.  Its ratio of profits to total capital (the equivalent of owners’ equity for a ‘for-profit’ private sector firm) ranged from around 100% to 150% over the same time period. See charts on pages 7 (ROE) and 11 (ROA).

 

How has the FED been able to achieve such a high ROA?  Using the language of financial theory, it has a very low weighted average cost of capital or WACC.  It pays virtually nothing on many components of its capital.  Until very recently, reserve deposits of member institutions were paid zero interest on them.  That was recently changed but still is now at only a fraction of 1%.  In past years a number of services were provided free or at very low cost to the member institutions by the FED.  But gradually in the Post WWII period, those services became costlier to the member institutions. 

Another very large source of debt capital is the issuance of Federal Reserve Notes.  This of course comprises nearly all of the legally circulating paper money of our system.  The Notes are a non-redeemable, non-interest bearing liability of the FED, giving this component of its capital a zero cost.  Some of the liabilities such as U.S. Treasury Deposits earn a more market like rate but are a relatively small portion of its liabilities. 
 

A peek at the FED as the federal fiscal...

http://www.federalreserve.gov/publications/annual-report/2010-federal-reserve-banks.htm#3 







Fed as Fiscal Agent.jpg




For a more in-depth look at the FED’s role as fiscal agent…


The Federal Reserve Banks as Fiscal Agents and Depositories of the United States

http://www.federalreserve.gov/pubs/bulletin/2000/0400lead.pdf  







FED Balance Sheet H.4.1 Total Liabilities and Assets.jpg









FED Financial or Operating Leverage (Liabilities over Assets).jpg




Finally, the ‘Capital Paid In’ receives a dividend of 6%.  This results in a very low WACC.  This means that the margin between the earning assets of the FED and the cost of capital on the liabilities and capital that finance the assets is very large compared to private sector firms’ margin that have a relatively higher weighted average cost of capital or WACC.

These are much larger leverage factors the FED experiences compared to commercial banks, for example, and as result the FED earns much higher return on equity or ROE than do private sector ‘For-Profit’ firms.


Some further articles to read related to the Fed



IS THE FED TIGHTENING THE SCREWS ON BANK LENDING? 
…risk based capital ratios reconsidered

CAPITAL RATIOS, OPERATING AND FINANCIAL LEVERAGE


July 16, 2011

http://www.econnewsletter.com/73901.html  


WHAT PRICE BAILOUTS? WHAT PRICE THE ELIMINATION OF THE GLASS-STEAGALL RESTRAINTS AND THE continuance of the ‘TOO BIG TO FAIL DOCTRINE’?

September 30, 2011

http://econnewsletter.com/92301.html  
 



WHAT IS LEFT FOR THE FEDERAL RESERVE POLICY MAKERS (FED) TO DO? 

Could a QE 3 be meaningful?


August 8, 2011

http://www.econnewsletter.com/82301/index.html


 

Why are interest rates so low?

January 12, 2012

http://econnewsletter.com/110101.html    


(The Financial Fiasco of Two-Thousand Eight (FFTTE))

January 2, 2009

http://byrned.faculty.udmercy.edu/2009%20Volume,%20Issue%201/2009volumeissue1.htm




THE FEDERAL RESERVE SYSTEM

December 31, 2003

http://byrned.faculty.udmercy.edu/2003%20Volume,%20Issue%205/newsletter%20fiveA.htm


...there are many more on the FED as well

http://byrned.faculty.udmercy.edu/




------- End of Part I of this series -------