Monetary_System

Overview of the U. S. Monetary System

                        Scott Buchanan    

                         May, 2013

Abstract:  This paper outlines the fundamental operations of the U.S. monetary system, which is representative of most other developed nations outside the eurozone. The nature of money, and the interactions of the Treasury, the Federal Reserve, and commercial banks are described.

Today’s fiat monetary system differs in several important respects from the gold standard that prevailed before the 1930’s. Economic textbooks have not fully caught up with these changes. The money multiplier, IS-LM, and loanable funds models are still taught, but these models do not correspond to current operational realities. Banks do not require reserves or prior deposits to extend loans; rather, it is loans that create deposits.

Analysis of sectoral balances shows that persistent government deficits can be a responsible policy. The role of trade deficits and currency manipulation on domestic employment and budget deficits is discussed. Government deficits are currently funded by issuance of debt securities, but debt-free financing (as with the Civil War greenbacks) is an option which deserves consideration.

Contents

What is Money?

Measurements of Money Supply

Clashing Schools of Economics

Monetary Realism: Economics Without Politics

Basic Operations of a Fiat Monetary System

Money Creation in the Private Sector

   The Myth of the Money Multiplier and Fractional Reserve Banking

   Bank Lending is Not Constrained by Deposits or Reserves 

   What Limits the Size of a Bank’s Loan Portfolio?

   Why Does It Matter How Money Creation Takes Place?

Sectoral Balances: Private, Government, and Foreign Surpluses and Deficits Must Net to Zero

   Applications of Sectoral Balances; Public Deficits and Private Savings

Historical U.S. Sectoral Balances

Trade Deficit and Government Deficit

   Chinese Currency Manipulation

Financial Assets and Real Wealth

Accounting for Investment and Savings

   Prior Savings Do Not Fund Investment; Rather, Investment Creates Savings

   Effects of Government and Foreign Sector Deficits on Private Savings

Federal Government Borrowing Operations

   Federal Borrowing via Bond Auctions

   Is Quantitative Easing “Printing Money”?

   Monetary vs. Fiscal Policy

   Inflation as the Real Limit on Government Deficit Spending

Can the Government Print Money? 

   Civil War Greenbacks: A Money-Printing Success Story

   The Trillion Dollar Coin Controversy

Should the Government Print Money?

   Printing New Money versus Retiring Debt

   Is Hyperinflation a Danger?

Is the National Debt Really a Problem?

   The Debt Ceiling and the Fiscal Cliff

   Japan: Pioneer of Astronomical Public Debt

   Greece and Other Eurozone Countries

 Goverment Fiscal Policy: Path to Prosperity or Road to Serfdom?

 * ** * ** ** * ** ** ** * ** * ** ** * ** * ** * ** * ** * * ** ** * * * * **

What is Money?

A voluntary trade typically results in enhanced value for both parties. Trades of goods and services can occur without money. For instance, a man who has a horse but would prefer a cow can exchange with someone who has a spare cow, and they can both come away happy with the deal.

However, in modern countries most large transactions utilize money.  Wikipedia offers the following definition of “money”:

Money is any object or record that is generally accepted as payment for goods and services and repayment of debts in a given socio-economic context or country. The main functions of money are distinguished as: a medium of exchange; a unit of account; a store of value.

To illustrate why all advanced societies use money, let us consider a village which operates on a barter system. (Many primitive cultures probably operated on a basis of mutual gifting rather than strict barter, but that does not affect the point at hand). Suppose the baker has produced two loaves of bread that will go stale in a day, so they would be best sold today. The baker would like a pair of sandals from the cobbler. They agree this is a fair trade, but the cobbler is backed up and will not be able to make the sandals for a couple of days. Thus, they cannot immediately make a swap. For this mutually advantageous transaction to proceed, the cobbler must agree to the obligation to deliver something (namely, a pair of sandals) in the future, in return for the loaves of bread today. The baker must believe that the cobbler will deliver on his promise.

Here we see the importance of credit and debt to enable transactions where there is a time lag between the actions of the two parties. The cobbler’s promise may just be the matter of a verbal agreement, but now suppose the cobbler writes down on a piece of paper, “The holder of this certificate is entitled to one pair of sandals from me, the village cobbler,” and gives this to the baker. Now this piece of paper is nearly the equivalent of a pair of sandals in value. It has become a store of value. If the baker later decided that he would rather have some candles instead of the sandals, he might be able to exchange this certificate for the candles. Thus, this piece of paper, which started as a statement of a debt between two individuals, would serve the function of money. (In fact, such “bills of exchange” issued by merchants were an important form of money in late medieval Europe).

Now suppose a farmer has a cow that he is ready to part with. All he wants at the moment is a pair of shoes from the cobbler. However, a cow for two shoes is probably not an acceptable bargain, and maybe the cobbler has no immediate use for a whole cow. This simple trade would probably not happen, since their wants do not coincide. To make this transaction work out, the farmer and cobbler may have to negotiate with the butcher (who is equipped to convert the cow into steaks), and also get other village craftsmen to promise to give the farmer some goods in the future in return for some of those steaks today. The farmer would have to weigh whether these craftsmen would keep their promises.  This becomes a big negotiation project, with the potential for gouging or inefficiencies due to all these links in the chain.

It would be much cleaner if the farmer could simply sell the cow to the butcher for something that functioned in that village as “money”, such as silver coins. The farmer could then use some of the coins to buy his shoes and retain the rest to buy other things later. In this case, the coins operate as a medium of exchange and also as a store of value. Both functions enormously aid financial transactions, which is a reason that money is so ubiquitous. Money also facilitates taxation by central governments, so governments have an incentive to put a monetary system in place.

For this system of money to work, the key players all have to believe in the value of the silver coins. Thus, money is a mainly social construct, an article of mutual faith. If people lose faith in the value of some form of money, it will in fact become valueless. For instance, during the American Civil War of 1861-1865 both the North and the South issued paper currency to fund their military efforts, since neither side had enough gold to pay their soldiers and suppliers of military goods. These were both fiat currencies, not at the time redeemable in gold.  Towards the end of the war, with excessive printing and with defeat of the Southern Confederacy in sight, the Confederate dollar lost nearly all its value. People no longer wanted to accept Confederate dollars in exchange for real goods, since they (rightly) feared that they would be unable to exchange the “greybacks” for the same amount of real goods in the future.

Sea shells were commonly used as money by native peoples in coastal areas of the Pacific and Indian Oceans, and the Atlantic coast of North America. Silver and gold coins were used as money in the Middle East and Europe for thousands of years. In the fourteenth century, Europe dropped silver coinage in favor of gold. Many older Americans and Europeans have a gut feeling that gold is “real” money, even though gold is just a yellow metal whose value is whatever people believe it to be.

In U.S. prisons, cigarette packs have historically served as money; since smoking in federal prisons was banned in 2004, cans of mackerel (available from the prison commissary) have become the preferred medium of exchange. The “digital currency” bitcoin was introduced in 2009 as a convenient, completely un-backed medium of exchange. It took several years to gain widespread acceptance, but usage took off in 2012.  By early 2013 the value of the bitcoin monetary base reached a billion dollars.

With the rise of banking in the Renaissance, banks issued paper certificates which were exchangeable for gold. For daily transactions, the public found it more convenient to handle these bank notes than the gold pieces themselves. In the late nineteenth and early twentieth centuries, leading paper currencies like the British pound and the U.S. dollar were theoretically backed by gold; one could turn in a dollar and convert it to the precious metal. However, the notion that we were really on a gold standard is just a myth: there was never enough gold in the British or American treasuries to redeem all the circulating money. Wikipedia notes, “In 1880, US government gold stock was equivalent in value to only 16% of currency and demand deposits in commercial banks. By 1970, it was about 0.5%. The gold standard was only a system for exchange of value between national currencies, never an agreement to redeem all paper notes for gold.”

Most countries dropped the convertibility to gold during the Great Depression of the 1930’s, so their currencies became entirely “fiat” money, not tied to any physical commodity. For the U.S. dollar, there was limited convertibility to gold after World War II as part of the Bretton Woods system of international currencies, but that convertibility ended in 1971. The value of fiat money derives largely from a government declaring that it must be accepted as a form of payment (“legal tender”) within that country, for “all debts, public and private”. The government typically requires that its citizens come up with some amount of its currency to pay their taxes, so that automatically creates some level of domestic demand for the currency.

Measurements of Money Supply

It is not straightforward to define what “money” is in a modern national economy. Simply tallying the amount of coins and paper currency is inadequate. Most buying and selling is now done by shifting numbers between abstract bank accounts, not by pushing a bundle of bills across a table.  Thus, these bank accounts serve the functions of money (medium of exchange and store of value). The question then arises as to which of these financial accounts to regard as money.

Among financial assets, there is a broad spectrum of liquidity. Typically you can write a check on your checking account which, when it clears, provides immediate and final settlement for a purchase.  On the other hand, if you want to tap your brokerage account with its holdings of Apple stock to buy a big screen television, you would typically have to sell (liquidate) your stock. A third party would have to be willing to give you something more money-like (e.g. credit your money market fund at your brokerage) in exchange for the stock at some negotiated price. Then you might have to transfer the funds from your brokerage fund into your bank checking account before you can actually buy that TV.  Because of all these intermediate steps, and the fluctuating value of the stock before you complete the sale, the stock holding would not be counted as “money”, even though its value enabled you to ultimately make your purchase.

The following are three common measures of money in the U.S.:

MB (“Monetary Base”): The total of all physical currency plus  Federal Reserve  Deposits (special deposits that only banks can have at the Fed). That is, MB = Coins + US Notes + Federal Reserve Notes + Federal Reserve Deposits. This is money essentially created by the government plus the Federal Reserve, which does not necessarily enter the private economy to be spent.

M1: The total amount of cash and coin circulating outside of the private banking system plus the amount of demand deposits, travelers’ checks and other checkable deposits. This is highly “liquid” money, i.e. accepted and used for transactions in the private economy.

M2: M1 + most savings accounts, money market accounts, retail money market mutual funds, and small denomination time deposits (certificates of deposit of under $100,000). The funds in these additional savings and money market accounts can in general be easily transferred to checkable accounts, and thus could go towards making purchases if desired.

The graph below from Wikipedia shows the growth in MB (labeled here as “BOGUMBNS”), M1, and M2 from 1981 to 2012. Prior to 2008, MB was mainly the paper currency, totaling less than a trillion dollars. Only about 5% of MB was reserves.  MB then ballooned to about 2.5 trillion dollars through the expansion of the reserve accounts, due to massive purchases by the Fed of assets such as mortgages and Treasury bonds in the wake of the financial crisis. (See here for details of MB composition over time.)  M1 has grown from about 1 trillion to over 2 trillion in the past decade, while M2 shot up from about 5.5 to 10 trillion dollars. All measures of money grew during the 2008-2009 recession, whilst economic activity as measured by GDP fell.  Thus, the velocity of money circulation through the economy dropped sharply during the recession. After 2009, money velocity has continued to drop, as the money supply increased while GDP was nearly stagnant.

 

 

Clashing Schools of Economics

It is striking how much passionate disagreement exists among experts who have devoted years of study to subjects like trade, employment, and deficit spending.  There are many opinions and schools of thought in economics. British economist John Maynard Keynes offered an analysis which promoted active government policies (including massive deficit spending) to tame the boom/bust business cycle.  His work has spawned “Keynesian”, “Post-Keynesian”, “Neo-Keynesian”, and “New Keynesian” approaches in the past half-century.

The Austrian school (representatives include Friedrich Hayek and Ludwig von Mises) remains influential, as does the Chicago monetarist school associated with Milton Friedman. The Austrians and Chicagoans are more leery of government intervention in the markets. Modern Monetary Theory (MMT), an outgrowth of chartelism, brings fresh insights into actual monetary operations in modern economies. This lecture by Missouri professor Stephanie Kelton provides an overview of MMT. In most cases, the theoretical understandings of these economics approaches are intertwined with advocacy for particular actions (or inactions) by national governments.

Proponents of these different schools bash away at each other in blogs and in the press. For instance, in 2009 arch-Keynesian Paul Krugman wrote an article in the New York Times Magazine assessing why the mainstream economics establishment failed to notice the approach of the global financial meltdown which started in 2007. There Krugman applied words like “crazy” and “absurdity” to the teachings of other leading economists (many of them Nobel laureates), and stood by his earlier claims  that “comments from Chicago economists are the product of a Dark Age of macroeconomics in which hard-won knowledge has been forgotten.” Ouch.

In response, Chicago economist John Cochrane dismissed Krugman as an “ex-economist”, since Krugman seems to spend more time writing viciously partisan newspaper columns than doing serious research.  MMT proponent  Bill Mitchell    in turn has dismissed all the other economic schools, claiming ”The frameworks they use to reason with are deeply flawed and bear no relation at the macroeconomic level to the operational realities of modern monetary economies.”

All this drama makes it difficult for the rest of us to have confidence that economists really understand what they are talking about. This is unfortunate, because the issues under debate have serious practical consequences.

The percentage of adult Americans with full-time jobs remains distressingly low.  The official U.S. unemployment rate is very slightly declining, but the numbers show that this decline is due largely to people giving up trying to find jobs, not from appreciable net job creation. Below is the Labor Force Participation Rate (percent of people of employable age who are working or actively seeking work), which continued to drop into 2012.

U.S. Labor Force Participation Rate . Source: Bureau of Labor Statistics

The same story emerges from the BLS release of the December 2012 employment figures. The formal unemployment rate edged down from 8.5% in December, 2011 to 7.8% in December 2012, yet the percentage of the total population with jobs was unchanged. The real unemployment rate (counting discouraged dropouts and involuntarily unemployed part-timers) is more like 14%, not 7.8%.

Americans are treated to the spectacle of the debt ceiling and the fiscal cliff. Europe and other parts of the world have their own economic woes. As a voter and a nano-scale investor, I would like to know whether increasing government spending can lead to sustained employment, and whether the ever-increasing national debt is a problem or not.

Monetary Realism: Economics Without Politics

Classical economics focuses mainly on the real economy, with less attention paid to the details of the financial world. Money is sometimes seen as merely a “veil” over the real economy, where goods and services are ultimately exchanged for other real goods and services. However, as was re-discovered in the crash of 2008, the operations of the financial world (banks, hedge funds, speculators, derivatives, re-insurers, etc.) and the level of private sector indebtedness can have massive real world repercussions.

In my readings in the area of economics, I have found a credible set of teachings that goes by the name of Monetary Realism . Monetary Realism (MR) aims to provide a description of the monetary system we actually have, without immediately jumping to policy recommendations. The hope is that as we understand the way things operate in the present regime, we can make better choices about how to manage or change it. MR is largely an offshoot of MMT, but with a better appreciation of private sector growth and production, and without MMT’s political agenda and abrasive edge.  Some key differences between Monetary Realism and MMT are described here.

As a window into Monetary Realism, I recommend “Understanding The Modern Monetary System” by MR co-founder Cullen Roche.   Roche is a veteran Wall Street professional who founded the Orcam Financial Group and who offers timely investing insights in his Pragmatic Capitalism blog.  Here I draw very heavily on the material in Roche’s paper.

Basic Operations of a Fiat Monetary System

The discussion here focuses on the U.S. finances, as representative of most other developed countries outside of the euro-zone. Two key actors are the U.S. Treasury, and the Federal Reserve System, which is the nation’s central bank. The Treasury is a department of the U.S. federal government. The Federal Reserve System (the “Fed”) is a group of twelve regional banks which was created by an act of Congress in 1913.  The Fed is basically a private corporation, owned by the commercial banks and operated by bankers, and largely serving the interests of the banks. On the other hand, it has some aspects of a federal agency, and is supposed to serve in the national interest.   The Fed has the power of creating money out of thin air, and is the issuer of U.S. paper currency (“Federal Reserve Notes”).  In contrast, Treasury is mainly a user, rather than an issuer, of money. To finance government operations, Treasury collects receipts in the form of taxes and of bond sales. The buyers for bonds include a select group of banks called “primary dealers,” other banks and corporations, and the Fed itself.

Commercial banks maintain “reserve” accounts at the Fed. These accounts are used mainly to settle transactions among banks, and also between banks and the Fed itself. There is also a requirement that a bank maintains some reserves as a percentage of certain types of deposits, but as discussed below this is no longer a major feature of the reserve accounts with the central bank.

Although the Fed is not a governmental entity, it operates under various mandates from Congress. Two key, often conflicting, mandates of the Fed are to achieve high employment and to maintain stable prices.  Also, the Fed provides financial services to the U.S. government and helps maintain the overall heath and functioning of the banking system. It provides a stable, efficient means for private banks to transfer funds among themselves, using their reserve accounts at the Fed.  Since its inception, the Fed has served to stave off the periodic wide-spread banking collapses that were a fixture of the nineteenth century.  In the wake of the financial crisis of 2007-2008, the Fed moved aggressively to buy up non-governmental assets, including mortgages and corporate securities, in order to keep key banks and other financial entities from collapsing.

Fighting inflation involves raising interest rates to slow economic activity, possibly to the point of provoking a recession. Politicians typically don’t want to bear direct responsibility for such unpleasantness. A justification for maintaining the independence of the Fed from direct government control is that it frees the central bank to act as needed to enhance the long-term economy, even if its actions are unpopular.

The Federal Funds market consists of banks lending and borrowing overnight among their reserve accounts at the Fed. Prior to 2008 the Fed typically controlled this short-term interest rate in the Federal Funds markets by buying and selling short-term T-bills (mainly in the form of short-term “repo” agreements) in open market operations to adjust the total size of these reserves.  The size of these reserves was relatively small, so relatively modest open market operations could make a big percentage change in the reserves, which in turn had a material effect on the interest rate that banks would charge each other in the Federal Funds market.

Impacts of Quantitative Easing

The Fed also influences the longer-term interest rates by buying or selling bonds in the open market. Unusually large-scale purchases of bonds by the Fed are termed “quantitative easing” (QE). These purchases typically have the effect of lowering long-term interest rates.  The money paid by the Fed for the bonds does not enter general circulation, where it might bid up the prices for other goods and services. Rather, in open market operations the Fed simply credits the reserve account of the bank from which it bought the bond. The balance sheet of the Federal Reserve System is shown here.

Since 2008, the Fed has purchased large amount of securities from the banks, which has caused the amount of excess reserves to balloon. This would normally drive interest rates in the Federal Funds market to zero.  That would have certain destabilizing consequences (e.g. for money market funds), so in order to keep interest rates from dropping to zero, the Fed now pays interest to the banks on their reserve accounts. The interest on excess reserves (IOER) that the Fed chooses to pay now largely controls the Federal Funds rate.  This represents a seismic shift from historic central bank practice. Open market operations are now largely irrelevant to short-term interest rates.

The Fed also directly sets the interest rate it charges if banks come to borrow additional reserves at the Fed’s “discount window.” The Fed prefers to have banks borrow from one another rather than directly from the central bank, so the rate at the discount window is typically maintained slightly higher than the target Federal funds rate.

As of early 2013, the Fed is still in full QE mode. The latest round of easing involves the purchase of $ 40 billion per month of mortgage-backed securities and $ 45 billion per month of longer-term federal government securities, in an effort to keep interest rates low.  It will take many years for the Fed to unwind the effects of all these securities purchases. It does not seem realistic that some $2 trillion of securities can be quickly sold back into the private sector, even after the economy starts to recover.  But until those securities are sold or simply mature in place, the amount of reserves will remain too high for the Fed to use them as a means to control short-term interest rates. Thus, if it is deemed necessary to raise interest rates to combat inflation, the Fed will have to raise the rate it pays on excess reserves  (IOER).  Currently the IOER is only 0.25%, which is well below the interest received by the Fed on the securities it holds. The Fed returns to the Treasury nearly all of this profit.  At some point, this profit may turn to a running loss, if IOER payout exceeds the income on the remaining securities. However, according to The Economist:

The Fed would not actually need a taxpayer infusion. One of the perks of central banking is that it can print the money needed to pay interest. If that generates a loss, it creates an offsetting “deferred asset” on its balance-sheet, representing future profits it won’t have to send to the Treasury. The forgone profit would pale in comparison with total interest saved, higher tax revenue due to stronger growth and roughly $500 billion of profit that QE had previously made possible.

Most of the run-up in cash (excess reserves) at the Fed in the past few years has been by foreign banks. If the IOER were raised to even 2%, this means payments of over $20 billion per year to foreign banks, which may to ignite political controversy.

To this potential operating loss may be added potential capital losses, if the Fed has to sell its securities for a lower price than it paid for them.  The Fed has only a thin margin of capital, so it wouldn’t take much capital loss for the Fed to go nominally insolvent and potentially require a bailout.  This is one reason why many observers believe that the Fed will not try to sell many of the securities it has purchased, but will simply hold them to maturity.

Money Creation in the Private Sector

In MMT the main narrative around money is that it is a creation of a central government to facilitate taxation. Money is created and initially spent into existence by the government. The government then requires (under threat of jail or confiscation of property) that citizens return some of this money to the state, which can then continue to spend it on state-enhancing activities. This is a means to force some of the resources (especially labor) in the nation to be directed into activities desired by the government.

That is a reasonable starting point, but it is important to note that the vast majority of money in countries like the U.S. is not created directly by government or central bank operations, but is created in the private sector when commercial banks make loans.   When individuals or companies decide to take out more loans (including loans for cars, houses, or business investment), the effective money supply in the nation increases.

A typical scenario of how bank lending increases money might go something like this: Fred would like to add an enclosed back porch to his house, but doesn’t have the money in hand to pay a carpenter to build it for him. So the base case is no payment to the carpenter and no porch for Fred. However, Fred realizes he can go the bank and get a loan to pay for the porch. So he obtains a $20,000 loan from the bank, which first shows up as a $20,000 credit to Fred’s checking account. Fred writes a check for $20,000 to the carpenter, who in turn pays $10,000 to a lumberyard for materials and keeps the other $10,000 as his fee. The lumberyard is able to pay its workers for that day, and order replacement lumber from a mill. The workers spent their pay on various items.  The carpenter puts $5000 of his $10,000 fee in a savings account, and pays the rest to a car dealer for a used car.

The initial loan to Fred set off a chain of spending and economic activity, which would not have otherwise occurred. Fred has his porch, the lumberyard workers continue to be employed and supporting their local merchants, the carpenter gets a second car, and this money keeps ricocheting around until it gets drained away into stagnant savings, or is used to pay down prior debt. Although they are not aware of it, the lumberyard workers’ pay for that day came out of the debt incurred by Fred.

The granting of that loan created $20,000 of spending capability, i.e. money.  As far as the economy is concerned, that $20,000 did not exist as effective money prior to the loan. Thus, the money came into existence simultaneously with the debt associated with the loan. Fred received the capacity to spend $20,000 today, but in turn accepted the obligation to pay back this money, with interest. It is assumed that Fred had a stable income, such that he would in fact be able to pay back the loan in the future.

In general, increasing debt increases the money supply, and paying down debt extinguishes money. For simplicity, suppose Fred repays the $20,000 loan (with $2000 interest added) in one big lump, two years later. In that year, he will presumably spend into the economy something like $22,000 less than he would have otherwise. Thus, his paying down of his debt acts as a decrease in the circulating money.

In normal times, as one person is paying down his loan (and thereby shrinking the money supply), someone else is taking out a new and even larger loan, so total debt and the amount of money in circulation stays about the same, or grows somewhat. A feature of the 2008-2009 recession, however, was a big drop in consumer demand for credit; folks decided to pay down debts and not borrow so much money to buy stuff. The effect was a big drop in spending and thus in overall economic activity (GDP) and in employment.

Where was that $20,000 before Fred borrowed it? We might think that it was sitting unused in the bank vaults, just waiting to be borrowed. That turns out to be an incorrect picture of the lending process, as discussed below.

The Myth of the Money Multiplier and Fractional Reserve Banking

Economic textbooks usually present this process of money creation via bank loans in terms of a multiplier effect resulting from fractional reserve banking. It all results from, and depends on, some initial deposit into the bank. The classic narrative goes something like this: first, Bank A obtains a deposit from Person 0 of say $100, and places it in a reserve account with the Fed. If there is a 10% reserve requirement, the bank has to keep $10 in the reserve account to back up the deposit in case the original depositor wants to take it back out. However, the bank is allowed to lend out the remaining $90 to Person 1. Person 1 takes that $90 and deposits it in Bank B. Bank B can then loan out 90% of that $90, or $81 to Person 2, leaving $9 in reserve. At this point, the combined bank deposits of Persons 0, 1, and 2 total $100 + $99 + $81 = $280, all based on one initial deposit of $100.

This successive loaning can keep going on and on, with the ultimate existence of up to $100 x (1/0.10) = $1000 in deposits. The reciprocal of the fractional reserve requirement (here, 1/0.1) is known as the money multiplier.  In conventional economics thinking, the central bank can control the overall amount of loans outstanding (and thus, in large measure, the money supply) by growing or shrinking the monetary reserve base, or by adjusting the fractional reserve requirement. This is thought to be an important tool of monetary policy.

In fact, this classical understanding of the money multiplier is largely incorrect for the fiat monetary system we have today. A bank does not need to have prior deposits or reserves to make a loan. A bank makes a loan out of thin air, and later obtains reserves as needed.

Bank Lending is Not Constrained by Deposits or Reserves 

Scott Fullwiler describes the step-by-step accounting entries involved with a bank loan, e.g. the loan described above to Person 1 (e.g. Fred, in the example above). As a result of making the loan to Person 1, the bank creates a deposit for Person 1 at the bank. In terms of standard double-entry bookkeeping, the bank records the loan as an asset, and the deposit as a liability, as shown in Figure 1 below. For Person 1, the loan is a liability and the deposit is an asset. Note that this can be done with no prior deposits or reserves. Thus, “loans create deposits”, not the other way around. This can be a difficult concept to grasp at first.

Suppose Person 1 buys a new car with the proceeds of his loan, and the car dealer deposits this money back in Bank A. From the bank’s point of view, the liability of the deposit owned by Person 1 is now replaced by an equal-sized liability to the car dealer (i.e. the car dealer’s deposit). Money has been created and spent, without the bank having any initial deposit.

If the bank needs reserves to support a deposit (either to meet a fractional reserve requirement or to cover the transfer of this deposit to a different bank), it can always borrow the reserves (at some interest rate) from other banks in the Federal Funds market or from the Fed’s discount window.  For instance, if Person 1 wants to call on the money deposited in Bank A as a result of the loan, he could withdraw that money and use it to pay a carpenter for a new porch, whereupon the carpenter deposits the money in Bank B.  Fullwiler’s treatment of the accounting is extended to the case of the transfer of the deposit from Bank A to Bank B:

For Bank A to settle with Bank B, Bank A at this point does need funds in its reserve account at the Fed, since that is where such inter-bank transactions are cleared. If Bank A had no reserves available with the Fed, its reserve account would go negative. However, as part of its mandate to maintain smoothly functioning markets, the Fed will temporarily allow an overdraft. This allows Bank A to go into the overnight markets, and borrow the funds (e.g. from other banks in the Federal Funds market, or from the Fed itself at the discount window) to replenish the reserves which were depleted by the deposit withdrawal. As long as the interest rate Bank A pays on these borrowed funds is less than the interest rate that Customer 1 is paying on the original loan, Bank A can make a profit here.

This is the opposite of the classical, money-multiplier view where prior deposits or reserves enable the loaning out of those deposits or reserves. If this classical view were accurate, the central bank or government could finely control the amount of lending (and hence money creation) by changing the amount of reserves and the fractional reserve requirement on deposits. This might have been true back in the days of the gold standard, but it is no longer the case. This control lever for the economy does not exist. Roche quotes a recent Fed paper which says:

Simple textbook treatments of the money multiplier give the quantity of bank reserves a causal role in determining the quantity of money and bank lending and thus the transmission mechanism of monetary policy.  This role results from the assumptions that reserve requirements generate a direct and tight linkage between money and reserves and that the central bank controls the money supply by adjusting the quantity of reserves through open market operations.  Using data from recent decades, we have demonstrated that this simple textbook link is implausible in the United States for a number of reasons. …  Changes in reserves are unrelated to changes in lending, and open market operations do not have a direct impact on lending. We conclude that the textbook treatment of money in the transmission mechanism can be rejected.  Specifically, our results indicate that bank loan supply does not respond to changes in monetary policy through a bank lending channel, no matter how we group the banks.

To recap on reserves, and their impact on lending by banks: Pre-2008, the Fed tightly controlled the amount of reserves (via open market operations) as a means of controlling the Federal Funds rate. It targeted a certain rate, and manipulated the reserves to support that rate. Post-2008, with the system awash in excess reserves, the Fed controls the Federal Funds rate largely by the rate it pays on excess reserves. In either case, the Federal Reserve would always provide plenty of reserves to any bank which is willing to pay the interest rate being targeted by the Fed.

The Fed’s manipulations do affect the total amount of bank lending, but only indirectly, via forcing interest rates higher or lower, which make bank clients less or more inclined to borrow. This borrowing/lending depends largely on the decisions of private individuals and institutions to borrow, and on private banks’ decisions to lend. Banks do not lend out their reserves. In terms of causation, the amount of lending by the banks determines the size of the monetary base, not the other way around. Tom Brown provides worked-out examples of banking transactions which illustrate various aspects of banking and money.

Nevertheless, it is still advantageous for a bank to attract deposits. In normal (i.e. pre-2008) times, the interest rate that a bank has to pay on the money borrowed in the Federal Funds market is appreciably higher than the interest rates the bank would pay its depositors.  A thorough discussion of the role that deposits play with respect to loans is here.

To further emphasize the unimportance of reserve requirements, we note that the 10% reserve requirement on deposits in the U.S. only applies to checking accounts which are held at large institutions. However, the vast majority of credit creation is in savings accounts, money market funds, and certificates of deposit.  For these instruments, the reserve requirement in the U.S. is zero. The reserve requirement on deposits in general has been disappearing in the Western banking system: it is zero in Canada, and is a miniscule 1% in the eurozone.

What Limits the Size of a Bank’s Loan Portfolio?

If reserve requirements are so unimportant in the overall credit markets, is there any limit on how many loans a bank can make? Can some tiny institution make a trillion dollars in fresh loans, then just run to the Fed and borrow (presumably at a lower interest rate than the fresh loans, in order to make a profit on the transactions) up to a trillion dollars in reserves to clear the transactions? The answer is that there is a different constraint, which is that banks are required to maintain a certain amount of secure “capital” (such as shareholders’ equity, government bonds, or other “tiers” of assets) to ensure that they can make good on their loans. This article provides an introduction to the subject. The rules here are subject to flux, interpretation, and evasion, but a typical constraint might be that a bank must possess capital in the amount of 8% of its loan portfolio.

In our current system, credit is rationed by the private, for-profit banks. The primary determinant of the amount of lending by a bank is not reserve or capital constraints, but whether the bank sees a sufficient profit opportunity to risk lending.  Even more important than the interest rate is the confidence that the loan will be repaid.  This incentivizes banks to lend against collateral or assets, as opposed to lending for investment in production. The result is that new money tends to be channeled into property and financial speculation rather than to small businesses and manufacturing.

Our focus here is on conventional banks, but it should be noted that a largely unregulated “shadow” banking system  has emerged in the last two decades which now rivals in size the commercial banking enterprise. The shadow banking system, which includes entities like hedge funds and money market funds, effectively creates money by its lending and provides much of the funding for businesses, but can contribute to systemic risk of collapse. This system has evaded regular banking regulation because it does not take in formal deposits, and thus does not merit federal deposit insurance.

Why Does It Matter How Money Creation Takes Place?

This loans-create-deposits issue is still misunderstood even among economics professors, and it does have policy implications. For instance, Krugman clearly failed to grasp this, and was roundly corrected by MMT proponents Scott Fullwiler  and Steve Keen.  Keen includes a quote from then New York Fed Senior Vice President Alan Holmes, who refers to the “naïve assumption that the banking system only expands loans after the System (or market factors) have put reserves in the banking system.”  Krugman was guilty of exactly that naïve assumption. According to Holmes, “In the real world, banks extend credit, creating deposits in the process, and look for the reserves later”. Keen’s article presents some important macroeconomic areas where one would be led astray by an incorrect understanding of how banks extend credit.

Misunderstanding of actual banking practice continues to lead some economists into fearing that the enormous level of excess reserves generated by the Fed’s quantitative easing constitutes a powder keg that might lead to an explosion of excessive credit and massive inflation. For instance, Austrian economist Robert Murphy in 2009 made an embarrassingly wrong public bet in that inflation would hit at least 10% by January, 2013.

The Positive Money blog puts it:

The money multiplier model of banking has several implications: Firstly, this model implies that banks have to wait until someone puts money into a bank before they can start making loans. This implies that banks just react passively to what customers do, and that they wait for people with savings to come along before they start lending.     Secondly, it implies that the central bank has ultimate control over the total amount of money in the economy. They can control the amount of money by changing either the reserve ratio or the amount of ‘base money’ – cash – at the bottom of the pyramid….Thirdly, it implies the money supply can never get out of control, unless the central bank wants it to.   Unfortunately, the money multiplier model of banking is completely wrong.

Another, somewhat pedantic, example of the importance of understanding the banking system is the “IS-LM” model (explained here) which appears in many economics textbooks as a starting point for understanding relations between interest rates and real outputs, and the financial market. The LM piece of this model assumes that the central bank can exogenously set the money supply. But our discussion above shows that the size of the money “base” created directly by the central bank, sometimes erroneously termed “high-powered” money, is a not a parameter that the central bank can independently fix.  Before 2008 amount of reserves was manipulated as needed to defend a target interest rate, while since 2008 the reserves have ballooned as a result of QE.  Oxford economist Simon Wren-Lewis confessed:

 I have become more and more embarrassed at having to teach the IS-LM model. The IS curve is fine, but the LM curve is not. The reason is obvious enough: central banks do not operate a fixed money supply policy. It would be nice to tell students that the fiction that the monetary authorities fix money is a harmless fairy story, but I do not believe this.

Sectoral Balances: Private, Government, and Foreign Surpluses and Deficits Must Net to Zero

Derivation of Sectoral Balance Equation

One way to define Gross Domestic Product (GDP) of a country is by summing up the costs of inputs into national production, via the following relation:

GDP = C + I + G + (X – M)

where C = consumption, I = investment, G = government spending, X = exports, and M = imports .

“Investment” here means mainly the purchase by businesses of goods like factories, machinery, and delivery vehicles that are not consumed immediately, but are used for future production of goods and services.  New residential housing is included in “I”, as is buildup of inventories, and on-the-job training of workers which boosts their productivity. (This economics definition of investment in (mainly) physical goods used to produce more goods differs from “financial investment”, which involves buying assets such as stocks which are hoped to increase in value. The word “capital” also is used in different ways in economic theory and in finance).

GDP can also be defined by the categories of the total spending of the national income:

GDP = C + S + T

where C = consumption, S = saving, T = taxes.

These two expressions for GDP can be equated, with C dropping out, to yield the following sectoral balance equation which relates the net savings (i.e. accumulation of net financial assets) among the private, public, and foreign sectors:

(I – S) + (G – T) + (X – M) = 0

where (I – S) = private domestic sector balance, (G – T) = public sector balance, and (X – M) = foreign sector balance. The foreign sector balance is also called (with some important differences in meaning among these terms which we gloss over here) the “national account”, the “current account”, the “capital account”, or the “trade balance”. The three sectoral balances must and do always sum to zero over a given time period. This is simply an accounting identity, not an economic theory. However, the mechanistic causality of how changes in one sector can drive changes in other sectors is subject to study and debate.

Applications of Sectoral Balances; Public Deficits and Private Savings

This equation can be rearranged in several ways, to emphasize differing relations. For instance, here we isolate the private domestic sector balance:

(S-I) = (G-T) + (X-M) ,     i.e.

(private sector net savings) = (government deficit) + (trade surplus)

This relation is important in understanding the role of government surpluses and deficits. If we for the moment ignore the foreign sector balance, this says that in order for the private sector to accumulate net financial assets, the government MUST run a deficit.  This can be difficult to accept at first, since it seems to justify “irresponsible” government budgets.

Although bank loans bring money (i.e. liquid, spendable deposits) into existence, banks cannot increase net financial assets. In standard double entry accounting, for every asset created, there is a second asset diminished, or a corresponding liability created. When a bank creates a spendable deposit (“money”) for a customer through extending credit, it also creates an obligation for the customer to pay back that money plus interest. The period of money creation lasts only until the loan is repaid, at which point that particular bit of money disappears.  When the Fed adds to a bank’s reserve account balance, that does not increase the bank’s net assets; in order to receive these reserves, the bank typically must (details vary) give or loan to the Fed some other asset, or take on the obligation of paying back the reserves.

Another way of thinking about this is to consider what would happen if the government ramped up taxes and ran large budget surpluses, year after year. At first blush that would seem like responsible governing, but that would eventually suck all the money out of the private sector. Assuming the amount of real assets did not change much, there would be deflation as less money was available to purchase goods and services. In general, deflation is bad for economies, as people are incentivized to hoard cash instead of spending or investing in productive capacity, and it becomes hard to pay off loans. Thus, we need to be careful not to assume that what is sound policy for a business or household necessarily carries over to government finances.

Here is one more, highly artificial example to help convey this concept.  Imagine an island economy where one-dollar bills printed by the central government were the only form of money (no banks or merchants’ certificates).  The only way these bills can enter circulation is for the government to issue them. Typically, the government would not simply hand them out for free, but would issue them in return for some goods or services. This would be “deficit spending”, since these bills had not been collected beforehand via taxation. So the money in circulation in the economy first gets there via government deficit spending.

Now suppose that, after an initial period of deficit spending which put a certain number of coins in circulation, the island government ran a balanced budget year after year. This would maintain a fixed number of bills in circulation, which might work well if the number of people on island, and their productivity, remained unchanged. But if the population increased, and improved means of farming and other production were devised, then a greater amount of goods would be chasing a fixed amount of money, and deflation would result. To maintain steady prices, the government would have to undertake additional deficit-spending in order to inject more financial assets into the economy. If the citizens wanted to save 5% of their annual incomes and stash those bills under their mattresses, then the government would have to issue the same number of bills into circulation in order to stave off deflation.

If we now consider the impact of the foreign balance, the equation above indicates that if the government ran a balanced budget, then a trade surplus would be necessary in order for the domestic private sector to experience net savings. A trade deficit would drain their financial assets, i.e. force S below I. Indeed, the steady decline in the U.S. personal savings rate since 1980 roughly  parallels the growth in the trade deficit.

(S-I) represents the accumulation of net financial assets across the whole (households plus business) private sector.  If over some time period (S-I) were to equal zero, that would not mean that households and businesses are saving nothing, it just would mean that their savings happen to exactly equal the business investment. Historically, households have wished to save an aggregate amount which is larger than the aggregate business investment (corresponding to S > I), and government deficit spending has provided a means to accommodate that desire to save, and to inject increasing amounts of money into the economy to provide for increasing GDP.

Another way to arrange the sectoral balances is:

(G-T) = (S-I) + (M-X) ,  i.e.

Government Deficit = Private Domestic Savings + Trade Deficit

This formulation shows that a push by the private sector to pay down its debts will tend to translate into a public deficit. Also, nations with a big trade deficit tend to also run a big public sector deficit. We will see below that a key component of the U.S. government fiscal deficit in most of the past two decades was the foreign trade deficit. Conversely, net exporters like Persian Gulf oil producers, Switzerland, and pre-2007 Canada have typically run government surpluses. This is not necessarily due to enlightened governance in these countries, but is mainly an outworking of the inexorable accounting identities.

Historical U.S. Sectoral Balances

The sectoral balances in the U.S. economy are shown in the two graphs below, taken from Roche’s paper. The first plot, covering 1952 to 2011, is formatted to show clearly the rigorous balancing of the three sectors for each time period. Here, the foreign sector balance is called the Capital Account. The second plot (1962-2010) shows the years more clearly, and has the foreign balance inverted compared to the first plot.

For most of these years, the private sector has run a surplus in the range of 1-4% of GDP, with a corresponding 1-4% government deficit. During and immediately following recessions (e.g. 1974, 1981, 1991, 2002, and 2008) there was typically a big spike up in the government deficit (as revenues slowed and welfare spending increased) , with a corresponding increase in net private savings. The meaning of “savings” or “S” here differs from our intuitive notion of individuals choosing to set aside a portion of their incomes. As discussed below, it turns out that net private savings across the whole economy cannot be produced by people trying to save. Rather, aggregate savings are a residual of decisions to invest, plus the effects of government and foreign sector surpluses or deficits.

The movement of the private sector into net dissaving in the 1998-2000 timeframe was an anomaly. This dissaving has been attributed to accelerated investment and increased willingness to take on debt in the euphoria of the internet/stock market boom, but that is not correct.  Decisions to borrow or invest or pay down debt do not affect (S-I). The negative value of (S-I) was due to tax increases and a generally robust economy which produced the first government surplus seen in decades.

At the time, the government surplus was widely hailed as a great accomplishment. However, it was also a symptom of growing fragility in the private sector.   Jamie Galbraith recounted how he was invited to the Clinton White House in April, 2000 to participate in a panel discussion of the current budget surplus. Most of the economists present regarded the surplus as a wonderful opportunity to pay down the national debt, cut taxes, etc. Galbraith was (literally) laughed at when he pointed out that if money was accruing in government coffers, it was being taken out of the hands of companies and individuals so it could not be spent or invested. “I said economists used to understand that the running of a surplus was fiscal (economic) drag,” he said, “and with 250 economists, they giggled.”  However, Galbraith’s concern was shown to be well-placed, as within a year the economy fell into recession.

The private sector again fell into net dis-savings in the 2004-2007, as the trade deficit ballooned due largely to high oil prices and to manufacturing shifting to China. The government deficit was not sufficient to offset the foreign deficit here.  In the same time-frame, private debt rose as people (thinking that housing values would keep going up forever, and enabled by predatory lenders) took out big mortgages for buying houses and also tapped their house values with home equity loans. This rise in household debt, while it did not directly cause (S-I) to drop, was a warning sign which was largely disregarded by the economics establishment (e.g. “Given the fundamental factors in place that should support the demand for housing, we believe the effect of the troubles in the subprime sector on the broader housing market will likely be limited, and we do not expect significant spillovers from the subprime market to the rest of the economy or to the financial system” – Fed chairman Ben Bernanke, May 17,  2007).

With the onset of recession in 2008, government deficits shot up to 12% of GDP and, just as suddenly (by accounting necessity), private sector net savings (S-I) recovered as government money poured into the pockets of individuals (e.g. via unemployment insurance payments) and busineses (e.g. massive bail-outs).  Nonresidential investment rose rapidly in the 2003-2007 timeframe, but collapsed by more than 20% in 2008-2009. Investment has risen steadily since 2009, and should reach pre-recession values by 2014.

The graph below shows the steep rise in household indebtedness (including housing mortgages) in the years prior to the Great Recession, and a remarkable, sustained deleveraging since then. The vast majority of that 13-14 trillion dollar debt consists of housing mortgages.  As households have continued to reduce their debt loads (e.g. by not buying expensive houses and by “losing” their mortgage debt through foreclosure or deliberate default), government deficits have hovered around 10% of GDP.  It looks like it will take several more years for this deleveraging to work household debt loads down to a more manageable level; only then can normal economic growth resume.

Non-mortgage household debt is broken into “revolving” (credit card-type) debt, and “nonrevolving” debt (mainly education and automobile loans). The revolving debt dropped from about $1 trillion in 2008 to about $ 0.85 trillion in 2010, and has held fairly steady since then. Nonrevolving debt stayed flat at about $1.5 trillion during the 2008-2009 recession, and then resumed its upward march, fueled mainly by increased college loans.

Trade Deficit and Government Deficit

As noted earlier, a foreign trade deficit tends to accompany a deficit in government budgets. The sectoral balance graphs above show that since 1984, very roughly and on average, the size of the U.S. trade deficit has been about half the size of the government deficit. If all else could be held constant (which is admittedly unrealistic), this suggests that if the trade deficit were eliminated, the U.S. government deficit might be reduced significantly.  How this might be accomplished is not so clear.

The American dollar serves as the world’s main “reserve” currency, i.e. as a standard basis for many transactions. For instance, if a South African refinery wants to buy oil from Saudi Arabia, they cannot do it in South African rand. Rather, they must first obtain U.S. dollars (in some electronic form), and deliver those dollars to the Saudis. This arrangement provides some marginal advantages for the U.S., but the extra demand for U.S. dollars contributes to the trade deficit in the amount of around 1-2% of GDP.

Free Trade Versus Protectionism

One issue that economists nearly all agree on is that hindering free trade, e.g. by imposing tariffs on imports, decreases total wealth. This concept is illustrated in Joy Buchanan’s Economnomnomics tutorial web site, where a two-island economy is described. On one island, conditions favor making crepes, and on the other island, Nutella brand chocolate hazelnut spread (which goes well on crepes) is manufactured more efficiently. Both islands gain if they exchange some crepes for some Nutella.

That said, there are national considerations beyond merely maximizing today’s goodies. Countries like Japan and Switzerland which are disadvantaged in food production may implement policies which retain some in-country agriculture, even if it costs consumers more than if they imported all foodstuffs from abroad.   Other governments take protectionist measures to maximize in-country employment, even though that reduces the amount of goods that its population enjoy consuming. In idealized economics this does not seem rational, but those governments are probably looking ahead to the results ten years down the road:  if protectionist measures continually boost in-country production, then in time, production will get large enough that in-country consumers will have more goods to claim, even if their consumption share of the production remains modest.  This strategy seems to have worked well for a number of Asian nations. When one country implements protectionist policies, it creates a dilemma for other nations: should they take compensating actions, or not?

A classical economics point of view is that trade deficits must correct themselves before too long. In the 1700’s David Hume argued that if England kept running a trade surplus with, say, France, then (under the prevailing gold standard), gold would be drained from France and accumulate in England. This would have the effect of causing prices in England to rise and prices in France to fall, until English goods were no longer cheaper than French goods.

This analysis assumed that (1) the excess gold in England enters the general money circulation, and (2) it can only be repatriated to France as payment for the purchase of French goods. These conditions may have held in the eighteenth century. If, however, English merchants were able to send this money back to France as investment (e.g. to purchase long-term assets in France, such as factories and housing) or as lending to French borrowers, that would short-circuit the self-correcting mechanism. A sustained trade imbalance typically involves repatriation of funds to the deficit country in the form of economic investment (buying physical assets) or as financial investment (buying bonds or other financial instruments that represent a claim on future production).

In our example, sending the gold to France in the form of investment would keep the English workers employed, and bring increasing quantities of French assets under English ownership. Whether this arrangement was good for France would depend on the uses of the English investment funds. If these funds were used to build improved factories in France, that would employ more French workers and grow the French economy. It would be less benign if the English merchants mainly took possession of existing assets (crowding out French buyers), or lent money so the French could go into debt to keep buying English goods.

The latter scenario is being played out in the U. S. housing market in 2013. In some areas of the country, as soon as houses come on the market they are snapped up with all-cash offers from investors. This makes it harder for working Americans, who typically need to go through the time-consuming step of obtaining a mortgage, to buy a home. Much of this crowding-out  investment money comes from abroad, deriving from Asian trade surpluses with the U.S.

Chinese Currency Manipulation

A leading example of current protectionism is China’s manipulation of its currency to keep it undervalued. This keeps Chinese goods cheap to outsiders, and supports Chinese production and exports. The figure below outlines how this is done.  The U.S. private sector sends dollars to buy merchandise from Chinese manufacturers.  The Chinese government directs the private sector to give many of those dollars to the Chinese central bank (People’s Bank of China or PBC). The PBC creates renminbi to give in exchange. This allows them to largely control the exchange rate of the renminbi versus the dollar. (“Renminbi” is the overall name of the Chinese national money, abbreviated “RMB”; the main unit of denomination of renminbi is the “yuan”).

Normally, this increase in circulating renminbi would lead to an increase in price levels in China. Such internal inflation, combined with a more or less fixed currency exchange rate, would tend to make Chinese goods more expensive to the outside, and thus naturally even out imbalances in trade. This is what normally happens when a small country pegs its currency to that of large trading neighbor.

However, the PCB takes the additional step of issuing interest-bearing bills or notes to soak up or “sterilize” the RMB it gives the private sector for their dollars. This restricts the amount of circulating currency for the populace to use.  The government also implements measures that restrict lending to consumers. Thus, the Chinese government policies quash internal consumption, which in turn promotes savings and reduces imports.

China largely sends those dollars (directly or indirectly) back to the U.S. to buy U.S. Treasury bills, notes, and bonds. This has some effect in keeping U.S. interest rates low, but we note elsewhere the Fed is capable of pinning both short-term and long-term as it wishes, without help from foreign investors. Government spending puts extra money in the hands of U.S. consumers so they can continue to buy imported Chinese merchandise.

A dollar of trade deficit means a dollar decrease in U.S. GDP, with concomitant loss in employment and production. The size of the U.S. budget deficit is increased by some substantial amount by the trade deficit, since the lost employment both reduces tax revenues and demands more government spending on relief.

Dollars circulate counterclockwise in this diagram. The net transaction is that U.S. Treasury debt is exchanged for the imported goods. Thus, the real cost of the lawn furniture and flat screen TV’s bought at Walmart may largely be paid by future American taxpayers.  This is a simplified and incomplete presentation of a complex set of transactions, but identifies some key issues.

Opinion varies on how severe a problem this is. Some advantages to the current situation are that American consumers have access to a flood of cheap goods, and American businesses profit from this international trade.  The Obama administration’s approach is that China has already let its currency appreciate a bit over the past few years, and that it is close enough to fair value to live with. Others disagree; the issue is not merely the currency exchange rate, but a whole complex of policies which inhibit imports and subsidize exports.    A recent study by the Economic Policy Institute concluded that eliminating global currency manipulation could create up to 4.7 million U.S. jobs and shrink the federal deficit by up to $ 165 billion.  Peter Morici wrote in 2012:

Faced with mounting difficulties in its real estate market and banks, Beijing is boosting tariffs and putting up new barriers to the sale of U.S. goods in the Middle Kingdom. President Obama has sought to alter Chinese policies through negotiations, but Beijing offers only token gestures and cultivates political support among U.S. multinationals producing in China and large banks seeking business there….Cutting the trade deficit in half, through domestic energy development and conservation, and offsetting Chinese exchange rate subsidies would increase GDP by about $525 billion a year and create at least 5 million jobs.

Besides the effects on U.S. employment, the trade deficit puts enormous sums of money in the hands of foreign investors. The global movements of this capital, estimated to be about $70 trillion, can cause inadvertent booms and busts. The Economist observed, “The trade deficit, less than 1% of GDP in the early 1990s, hit 6% in 2006. That deficit was financed by inflows of foreign savings, in particular from East Asia and the Middle East. Much of that money went into dodgy mortgages to buy overvalued houses, and the financial crisis was the result.”

Some actions advocated to reduce the U.S. trade deficit are tariffs on imports, voluntary boycotts, or capital controls such as a tax on dollar-yuan conversions. Although Americans don’t normally think in these terms, it is not uncommon for nations to implement controls on the flow of capital across or within their borders. China, for instance, strictly limits the amount of foreign investment capital entering the country. For a nation like the U.S. which generally prefers free trade and free capital movement, capital controls seem like a blunter instrument than tariffs focused on the goods of just a few offending trade partners.

We have focused on the U.S.-China trade as the most spectacularly imbalanced example, but there are plenty of other examples of trade distortion, including American and EU agricultural subsidies. While currency manipulations exacerbate a trade deficit, the actions of individual citizens can play a major role. There is evidence that French and German citizens are generally aware that a positive trade balance can help sustain employment and production, and thus they are willing to make some personal sacrifices in order to buy domestic products and to facilitate export-oriented industry.

Financial Assets and Real Wealth

The discussion so far has dealt mainly with financial assets, rather than the store or flow of actual goods and services. While net financial assets can be created in the private sector if the government runs a sufficiently large deficit, that does not mean that government spending necessarily creates real wealth. It might just generate inflation, if more dollars start chasing the same goods.

A subsistence farmer can create eight chickens out of two chickens (representing an increase in real assets), with minimal interaction with the monetary world. No government deficit was needed. Real wealth is produced in the real world, although monetary operations can facilitate the deployment of resources to facilitate that production.

The amount of goods and services available per person for consumption is approximately equal to the total goods and services produced per person in that nation. Thus, on average, and neglecting international trade, the economic well-being of citizens is directly defined by their productivity. Investment is key to increasing production. Here we mean investment in the economic sense of paying to install new manufacturing machinery or to implement a more efficient service business, not in the sense of shuffling financial assets around in hopes of gain. Increased real goods and services are obtained as a result of sound investments in improved methods of production.

We might think about dairy farming as an example of the benefits of improved technology and organization. In the nineteenth century, milk mainly came from a human hunched beside a cow, hand-milking into a bucket which was then hand-emptied into a larger container. Herds were small, and manure was shoveled by hand. During the twentieth century, mechanical milking devices were developed, which greatly reduced the labor involved. In a modern milking parlor, dozens of cows are led in, milking devices are attached to the udders, and milk is pumped through pipes into sterilizing facilities and storage tanks. One farm family can now run hundreds of cows, and computer monitoring helps the farmer provide the optimal nutrition to each cow. Similar increases in productivity have taken place in other areas of agriculture.  The result is that food costs as a percentage of disposable income have dropped from 21% in 1950 to about 10% today.    Fewer farmers are needed to produce all that food, and today’s farmer leads a less-punishing lifestyle than a century ago.

This productivity improvement required mechanical ingenuity to devise better physical equipment, and education and enterprise on the part of farmers to implement new, expensive methods.    In some cases, a farmer could buy new equipment or redo his barn just using his retained savings. If he lacked enough money to do this on his own, however, the financial market (e.g. his local bank) could provide a loan to enable him to make these investments. If all went well, the additional production and cost savings from the improved milking equipment would allow the farmer to pay back the loan and enjoy greater earnings.

While financial transactions by themselves may not create real wealth, they can enable resources to be directed to productive uses that will increase the availability of actual goods and services. A key issue is how much investment is needed. It is often not clear at the time how much total investment, or what specific investments, are appropriate.  Some investment is essential to growing wealth, while malinvestment, such as building a factory that is shut down soon thereafter, wastes resources and can contribute to a boom/bust cycle. Government policies can encourage investment (e.g. by tax breaks on spending for research) or hinder it by regulatory burdens and by erratic rulings that deny businesses the clarity needed to plan with reasonable confidence.  Likewise, private sector financial operations may facilitate useful real-world investments, or may merely serve to skim wealth into the pockets of clever manipulators. Thomas Jefferson [Letter to John Eppes, 1813] wrote, “the truth is, that capital may be produced by industry, and accumulated by economy; but jugglers only will propose to create it by legerdemain tricks with paper.”

We may note parenthetically that whether most citizens actually benefit from the increased average production depends on how the wealth is distributed. In the U.S. nearly all the increase in real income in the past 15 years has gone to the wealthiest 10 percent, such that the real median income has stagnated or declined.  The share of the national income accruing to corporate profits has risen steadily, with a corresponding drop in wages paid to workers. We assume that the wealthiest decile devotes a large share of its income to passive finance investment, rather than spending into the economy as consumption where it creates income for others. We speculate, therefore, that the increasingly skewed distribution of income is acting as a drag on GDP growth.

Accounting for Investment and Savings

Investment in improved means of production is crucial to raising standards of living. It is important, therefore, to have a clear understanding of how investment is financed. The sectoral balance equation can be rearranged yet again to yield an expression of where funds for investment come from:

Investment = Private Saving + (Government Surplus) + (Trade Deficit)

This seems to imply that, absent government or foreign imbalances, private savings finances business investment; and also that a government surplus and/or a trade deficit are positive developments for increasing investment, which in turn fuels growth.  These relations seem to contradict much of the discussion above, which painted a government surplus and a trade deficit as often harmful to the domestic private sector.

Fullwiler cautions that a simplistic interpretation of this equation is misleading, claiming that prior savings do not in fact fund investment:

The right-hand side of [this equation] is usually referred to as “national saving” by neoclassical economists, and more “national saving” is thereby thought necessary to raise more business investment in order to raise the economy’s long run capacity to produce goods and services. This interpretation is nearly ubiquitous in the profession.

However, this interpretation is inapplicable except for a fixed exchange rate monetary system operating under a gold-standard or currency-board type of regime in which there is in fact a “fixed” quantity of savings that exists or can be created. But in our flexible-exchange rate monetary system, saving does not finance spending; indeed, banks create loans without any prior deposits or reserves being necessary…Instead of saving to finance capital investment (which is not actually what happens), we have “private sector net saving,” which is the addition/subtraction to net financial wealth for the private sector in a given period… Thus, “national saving” as defined in the textbooks (private saving + government surplus + foreign saving) is a misleading concept in our monetary system, since if the government is “saving” some other sector (or combination of them) must not be, by definition.

In the physical world, savings do logically precede investment; the farmer must retain some of this year’s seed in order to sow a crop next year. In modern monetary workings, however, this is not necessarily the case. This can be counterintuitive, so below are given some simplified examples to illustrate how savings and investment are related.

If the government and foreign sectoral balances were zero, then investment would exactly equal domestic savings, i.e. I=S.  A textbook interpretation is that household savings come first, and then companies compete for those funds in order to pay for their investments. In this supposed “Loanable Funds” market, interest rates adjust up or down such that individuals are incentivized to save and lend the exact amount of money that firms are willing to borrow at the market interest rate.

Andy Harless points out that this is not how things actually work. There is no such funds exchange which, each quarter, matches to the penny what people put in the bank or under their mattresses with businesses’ investment schemes and inventory buildups.  Rather, the exact equality of savings and investment is an accounting artifact of how “savings” are defined.  It turns out that net savings in an economic unit are possible only if there is investment, as explained in Harless’s article and as illustrated below.  We are using the economics definition of investment as paying for some enduring asset, such as a factory building or machine, which has value in helping to produce goods or services in the future.

Here we are treating flow, not stock variables. We are concerned with transfers of money from one party to another during an accounting period, not how much money they had at the start of the accounting period. Thus, if you started the quarter with $1000 and did nothing with it, that would not count as net savings.

We commonly think of the act of saving as a matter of positive intent and action, such as putting coins in a piggy bank. In economics, however, savings are defined as “income not spent on consumption,” or “income minus consumption,” with no regard for intention to save. If you are a private individual and you receive wages for producing some good or service consumed by another individual, then your income will be matched (glossing over some details) by their consumption, and thus no net savings in the economy takes place.  On the other hand, if you receive income in compensation for producing some (typically physical) asset of enduring value (which is what we are calling “investment”), then the economy has income which is not offset by consumption. Only then can net savings be booked.

Examples:  Prior Savings Do Not Fund Investment, Rather Investment Enables Savings 

In Case 1 the PotCo company sells $1000 worth of pots to customers, and pays its workers $900 to produce these pots, starting from a clay pit that the company owns.  The $100 earnings are paid out as dividends to stockholders.  Total income ($900 wages for workers plus $100 dividends for shareholders) is the same as consumption ($1000 worth of pots purchased).  Since savings are defined as “income not spent on consumption,” or “income minus consumption,” the net savings here are $1000 minus $1000, which is zero.

How can net savings be produced in this situation? Suppose (Case 2, below) the customers, in an attempt to save money, decide to buy only $800 of pots. If the company is nimble enough to forecast demand correctly, it will lay off some workers to reduce its salary costs and produce only $800 worth of pots. The outcome would look something like this: 

The shareholders are still paid $100 in dividends, while wages have dropped to $700. Consumption is down by $200, but so is income, and thus there is still no net savings.  This illustrates the “Paradox of Thrift”: if many parties in a closed economy try to save by hoarding money and not spending it, this drives total incomes down, since one man’s consumption is the next man’s income. There is no net increase in savings, and everyone on average has fewer goods and services to enjoy.

In theory, the economy could adjust to having less money in circulation by an across-the-board reduction in both wages and prices, with real output held constant. In practice, however, such wage and price deflation often results in a slump in real GDP.   Households historically have desired to set aside several percent of their income per year, which pulls money out of circulation. Government deficit spending is required to inject counter-balancing money into the economy.

Suppose (Case 3, not illustrated) the company decides to retain its earnings instead of paying them out as dividends. In that case, the $100 in retained earnings is booked as income to the company. Total income is still $1000, and there is no net savings. The $100 sitting in the company’s vault is no more or less “saved” than if that $100 (as in Case 1) were sitting in the wallets of dividend recipients.

In Case 4, the company retains $100 in earnings as in Case 3, but also decides to pay a machinist $100 to assemble and install a durable machine which will facilitate future pottery production. This transaction is one of investment (purchasing a good valued primarily for its future value), rather than consumption. As a result of this transaction, the private sector now possesses an enduring asset that it didn’t own  before.

Total income is $ 1100 (= $900 workers’ pay + $100 retained earnings + $100 to machinist), while consumption is unchanged at $1000. Now, at last, there is a net savings in our “economy”, of $1100 – $1000 = $100, which is equal to the amount of investment. Thus, S=I. Note that this savings is a consequence of a decision to spend additional money to invest, not a result of a decision to hold back on spending.

In Case 5, PotCo (as in Case 1) pays out the $100 profit in dividends, but still wants to install the machine. In order to pay for it, the company borrows the money from a bank.  PotCo receives a spendable deposit, but incurs an obligation to pay it back, with interest. No income or consumption is immediately associated with the borrowing event itself. As in Case 4, savings are $1100-$1000 =$100, which is equal to the amount of the investment.

Again, there is a net savings as a consequence of a decision to spend on investment. Indeed, this particular investment was enabled by PotCo borrowing money and going into debt, which seems the opposite of our normal notions of savings.

Where did the bank get the $100 to lend?  A common assumption would be that some other parties (e.g. workers or shareholders or customers, in this picture) had earlier deposited $100 in the bank to allow it to lend.

However, as discussed at length above, banks do not need to lend out of existing deposits or reserves. A bank can make a loan, and obtain reserves later as needed to settle transactions. For short-term loans (and currently for longer-term borrowings), there is not a fixed, limited supply of funds. Rather, the Fed will make available, at its target interest rates, whatever reserves are needed to meet the demands of the banking system.

Suppose the bank borrowed directly from the Fed at its discount window to settle transactions arising from the money lent to PotCo and subsequently paid to the machinist. The economy would book $100 savings, there would be the $100 asset sitting in the PotCo plant, and the machinist would have an extra $100 income, all without some third party deliberately choosing to save. Nobody had to forego consumption in order to provide the funds to install this machine.

Normally the bank would have no need to borrow at the Fed’s discount window, since somewhere in the banking system there are enough deposits to back reserves for settling loans.  If consumers draw down their bank accounts to spend freely, the money they pay to businesses mainly ends up right back in the banking system as new deposits.  Again, there is no requirement for households to withhold spending in order to finance business investments. On the contrary, if consumers are hoarding money instead of spending it, that would discourage businesses from making the investments which would produce savings.

If a rigid gold standard prevailed, with no banking or credit, it might indeed be necessary to persuade folks to first give up some of their coins and forego consumption, in order to pay for installing an investment.  This is no longer the case for the financial system we have today. Not all economic modelers seem to have caught up with this shift.

This sounds almost too good to be true: consumers can spend freely, investments are seemingly financed out of nowhere, and savings automatically appear.  If this is so, why don’t we see frantic borrowing and investing, with consequent skyrocketing income and savings? The main answer is that, for sustained investing in the private sector, the investments must be sound ones. The amount of money saved, or extra money earned, as a result of making a physical investment must exceed the original price of the investment, in order to pay back the loan with interest. A viable investment is typically associated with an improvement in productivity, which allows more goods to be produced with less labor or resource cost.

There are only a limited number of prudent investment opportunities at any given time. Identifying them and acting on them in a timely fashion may well spell the difference between success and failure for a firm.

Both the company and the bank must be convinced that the investment will pay back. The company and the bank must come to agreement on how much interest will be charged on the loan.  Conventional banks are so conservative that for business production investments they may demand unacceptably high interest rates as a risk premium, or they may simply decline to make a loan at all.  Businesses may then turn to shadow banking entities like hedge funds or structured investment vehicles (SIVs) and negotiate for loans to fund their high risk/high reward ventures.  This competitive funding activity has some resemblance to the classical Loanable Funds market, although as noted above, this is not the explanation for why S=I to the penny for any reporting period. Also, through the rehypothecation  of assets (especially in the UK), the amount of funding that can be provided by the shadow banking system is in theory limitless, which does not fit the Loanable Funds model.

Effects of Government and Foreign Sector Deficits on Private Savings 

In Case 6, the government buys $200 worth of pots, without levying additional taxes. This is deficit spending, funded by bond sales or outright money printing. PotCo pays its worker an extra $200 to make these pots. This boosts private sector income by $200, without increasing household consumption. The extra $200 in private sector financial assets appears as a boost in savings, which occurred independently of any household decision to save. 

Case 7 is like Case 6, except that the shareholders spend their $100 dividend on foreign goods. This boosts consumption with no increase in household income, dropping private sector savings by $100:

S = I + (G-T) + (X-M) = 0 + (200-0) + (0-100) = $100

These examples all show that net private savings across the economy is completely independent of the desires or efforts of individuals to save. Rather, aggregate private savings is a residual of (mainly business) decisions to invest, government decisions on taxes and spending, and all the decisions that go into foreign sector deficits or surpluses.

Federal Government Borrowing Operations

Federal Borrowing via Bond Auctions

If the U.S. government ran a balanced budget, its expenditures would match tax collections. The norm, however, is for the government to spend more than it takes in. The Treasury works mainly through its accounts at the Fed.  The Treasury is by law not permitted to run an overdraft at the Fed, so it must come up with extra money by borrowing it via issuing debt instruments. These securities include short-term bills, intermediate-term (2-10 year maturation) notes, and longer-term bonds.  The purchasers of the securities effectively lend money to the Treasury to spend now, in exchange for an obligation of the U.S. government to pay that money back in the future, with interest.

The Fed manages the Treasuries auction process, working with the 21 large banks that have been designated “primary dealers.”  These banks are required to participate in auctions of Treasury debt, and in purchase the bulk of the auctioned securities.  The primary dealers then on-sell these securities to the public, including other banks.

A crucial point is that a Treasury bond auction can never fail.   As part of their contractual obligations to the Fed, primary dealers are required to make “reasonable” bids for the bonds. Of course, the market might demand a high rate of interest payment on the bonds, if there are expectations of inflation. As a last resort the Fed could buy up any unwanted T-notes, using its unlimited capability to create money. Thus, the U.S. need never default on its dollar-denominated obligations or otherwise go broke. The Fed/Treasury combination can always come up with more dollars, if needed.

Is Quantitative Easing “Printing Money”?    

As part of its open market operations to manage the Federal Funds interest rate, the Fed has historically bought and sold short-term Treasuries. Since 2008, the Fed has also bought longer-term debt as part of its quantitative easing (QE) efforts to reduce longer-term interest rates.  Some commentators claim that QE amounts to printing money, but that is not correct.  When the Fed buys a bond from a bank, it credits the reserve account of that bank.  From the bank’s point of view, it just swapped a long-term interest-bearing asset (the bond) for a short-term interest-bearing asset (the increased reserve account). No new financial assets are created in the private sector through QE, so it is not immediately inflationary.

The main way in which the funds from these reserve accounts would enter general circulation as “money” is if the banks used them to help support loans to some business or consumer who would then spend it on a house, car, or business expansion. However, since 2007 people have become less desirous of borrowing money, and banks have become more cautious about lending it. Thus, these expanded reserve accounts are largely just sitting inertly on the banks’ books.

A bank can use these excess reserves to purchase other investment-grade securities, such as highly-rated corporate or foreign government bonds.  In that case, the net change in the position of the bank would be in now owning this new security in place of the original bond which the Fed had purchased from the bank. So QE is largely a non-event as far as the immediate money supply and asset position of the banks.  QE does act to encourage lending, by lowering interest rates and providing abundant highly liquid reserves, but cannot of itself force an increase in the supply or velocity of circulating money.

However, as banks buy bonds from the public to then re-sell to the Fed, or as banks use their excess reserves to buy securities from the public, this does place additional spendable funds (money) in the hands of the public in place of the bonds. By this mechanism, a significant portion of QE funding has landed in assets that count as M1 and M2 money. This allowed M1 and M2 to climb during the 2008-2009 recession, even as bank lending (the normal source of money creation) collapsed.  QE has indirectly created money, but this money remains largely locked up within the investment community. Also, the Fed is not handing out free money here; households or pension funds are exchanging their bonds for other financial instruments that happen to be more readily spendable. Thus, QE involves “selling money”, not “printing money.”

Typically, the parties that currently own the types of securities (government and mortgage-backed bonds) which the Fed is buying as part of QE  are more interested in investing their money for returns, rather than spending it on consumption.  Thus, these extra funds injected into the private sector by the banks’ asset purchases  have mainly gone into financial investments (e.g., supporting the stock market), rather than into general circulation and consumption where they might boost employment and incomes. The Fed appears to believe in the “wealth effect”, whereby the good feelings engendered by higher stock prices will induce individuals to start spending more.

Monetary vs. Fiscal Policy

The Fed’s actions impact interest rates, which in turn affect overall economic activity.   This is termed “monetary policy”.  If the economy is overheated, the Fed can (normally, by restricting reserves) raise interest rates as high as necessary to cool it down. This sort of shock therapy was applied in 1980, when the prime interest rate was pushed up to 20% in a successful effort to tame inflation. If the economy is sluggish, the Fed can normally act to lower interest rates, making it more attractive to borrow and spend money.

However, once interest rates approach zero, this lever becomes ineffective.  This has been the case in the U.S. since 2008. Interest rates are as low as they can go, and plenty of liquid reserves are in the system, but customers have been reluctant to borrow, and banks have been reluctant to lend.  In Keynesian economics this situation is termed a “liquidity trap”, or “pushing on a string.” Under these circumstances, the most effective immediate means to stimulate economic activity is for the government to adjust its level of deficit spending. That is called “fiscal policy.” It is fiscal stimulus (deficit spending), not monetary policy like QE, that is effectively “money printing,” in the sense of directly injecting money into circulation.

Inflation as the Real Limit on Government Deficit Spending

The discussion here has touted the virtues of federal deficit spending, to provide financial assets to the private sector, and to maintain employment if private demand drops during a recession. There is no hard limit to how much the Treasury could borrow to finance this spending. This raises the question: why not massively deficit-spend all the time? Why not have factories humming, people employed, and goods flying off store shelves?

One answer is that excessive deficit spending would lead to undesirably high levels of inflation. If the added government spending pushes the aggregate demand for goods and services up against the productive capacity of an economy, typically prices rise sharply, as more money chases a limited supply of real stuff and government purchases compete with private consumption.  The increased spending would likely not lead to a further increase in employment or real production/consumption. Thus, the level of government spending should normally be curtailed if inflation starts to take off.

Businesses need predictable conditions in order to make optimal decisions on investment. Most economists agree that a steady but low (e.g. 2-4% per year) increase in prices is a good thing.    If the value of money steadily drops, it can encourage people to invest it instead of hoard it.    If workers in some industry are overpaid relative to the value they create, it is relatively easy to let their real wages adjust downward (if there is ongoing inflation) by simply not giving them salary raises. It is more difficult to directly cut nominal wages, which what would be required in a noninflationary environment.    On the other hand, high inflation (e.g. over 8%) can lead to hoarding of durable goods, devastation of pensioners living on fixed incomes, and perhaps set off a boom/bust cycle. (There are various theories which relate prices to real output, which are not discussed here). Opponents of central banking bemoan the drop in the value of a dollar in the hundred years since the Fed was established in 1913, but despite this accumulated inflation, real living standards for all Americans  dramatically improved in the past century.

In the recent Great Recession, massive amounts of deficit spending have been undertaken in an effort to stimulate the economy.  This led some commentators to issue dire warnings of inevitable inflation. However, five years into unprecedented peacetime federal deficits, significant overcapacity remains and inflation remains tamely below 3%.  This shows how slow the economy is, and also indicates that the government spending to date has not been excessive. Indeed, other commentators claim that the level of deficit spending should be increased dramatically in order to end the current stagnation.         

Can the Government Print Money?  

Government disbursements are done by transferring balances among accounts at the Fed. Under current law, the Treasury cannot spend funds unless those funds have been first been collected via taxation or via the sale of debt securities such as notes or bonds.   A consequence of ongoing deficit spending, then, is an ever-growing mountain of federal debt.  The payments of interest and principal on this debt get larger every year.

Why does this deficit spending have to be financed by borrowing? Why can’t the government actually print money? This could take the form of literally running the presses and cranking out stacks of $100 dollar bills to hand out, or by demanding that the Fed arbitrarily credit the Treasury’s reserve accounts with a trillion dollars, which the Treasury could then disburse. This would allow deficit spending to continue, but without piling up a national debt.

There is no physical or economical reason that the government cannot create debt-free money in these ways. The key reason this is not done is simply that the existing laws do not allow it. However, those laws could readily be changed. The U.S. Congress could pass legislation that empowered the Treasury to issue a certain amount of money, debt-free.

Civil War Greenbacks: A Money-Printing Success Story

This sort of money issuance by the federal government was practiced during the American Civil War.  It became clear that financing the war by taxation or borrowing from the banks was impractical.   After heated debate as to its legality, Congress passed laws authorizing the issuance of about $ 450 million dollars in United States Notes. These were not at the time exchangeable with gold, but they were designated as legal tender for private debts. Thus, a creditor who had expected to be repaid in gold could be repaid with these “greenbacks.” It is doubtful that the Union could have been preserved without this fiat currency.

The issuance of the greenbacks was originally seen as an emergency wartime measure. After the war, Congress voted to gradually retire them and return to a gold standard. By 1868, the amount of these Notes had been reduced to $ 356 million. At that time, however, the nation was sliding into a recession, and it was expedient to let the greenbacks continue to circulate. After some additional minor issuance and withdrawals, the amount of circulating U.S. Notes was fixed at exactly $346,681,016 in 1878, and stayed at that level for nearly 100 years afterward.

The legality of the greenbacks was challenged in several court cases after the war. U.S. Supreme Court decisions in 1871 and 1884 clearly granted the federal government the power to issue this sort of debt-free fiat currency.

While the American South suffered devastation of its industry and social order, production in the North soared to meet wartime demands. Although the greenbacks expanded the federal money supply, this was largely matched by an increase in real production, so runaway price inflation did not result.   Also, the greenbacks were only a portion of the North’s money supply.

The Trillion Dollar Coin Controversy

Buried deep within the United States Code is 31 U.S.C. 5112(k), which reads:

The Secretary may mint and issue platinum bullion coins and proof platinum coins in accordance with such specifications, designs, varieties, quantities, denominations, and inscriptions as the Secretary, in the Secretary’s discretion, may prescribe from time to time.

Although the original intent of this law was to allow the Treasury to mint small, affordable coins for collectors wishing to possess a platinum coin, this stature inadvertently gives the Treasury the authority to issue a coin in ANY denomination. This has given rise to the proposal that the Treasury mint a coin stamped “One Trillion Dollars”, walk it over to the Fed, and exchange it for a trillion dollars of outstanding U.S. Treasury notes and bonds. This would immediately drop the national debt by a trillion dollars, and (among other things) keep the debt below the existing debt ceiling. This would not be inflationary, since it would be simply retiring debt which was held by the Fed. No extra money would be put into public circulation.

This proposal got so much attention that an official reaction was given. In January, 2013, the Treasury issued a statement, “Neither the Treasury Department nor the Federal Reserve believes that the law can or should be used to facilitate the production of platinum coins for the purpose of avoiding an increase in the debt limit.” That was probably the right answer for this case. The near-term purpose of the trillion dollar coin would have been to sidestep the debt ceiling. If the Administration took such a momentous step in monetary issuance on the basis of a misapplied legal loophole, it would set off a bitter turf fight with Congress, which has Constitutional jurisdiction over federal spending.  None of this political posturing means that issuing a trillion dollar coin is economically unsound.

Should the Government Print Money?

The greenback episode established that the U.S.  government has the authority to issue debt-free fiat currency, and that a measured issuance of such currency can work out very well. What might this issuance look like today?

For simplicity, let’s consider two individuals, Jane and Joe. Jane starts with $1000 in a checking account. The government purposes to transfer $1000 into Joe’s account, via deficit spending. Under the current system, the government might sell a $1000 bond to Jane, in exchange for the money in her checking account. So after the dust settles from inter-bank transactions, Jane will own a $1000 bond, and Joe will have $1000 in his account, which he will likely spend on something.  The Treasury promises Jane to pay back her $1000, with interest, sometime in the future. In the meantime, her spending potential is crimped by something like $1000.

If the Treasury printed money without selling bonds, it would simply deposit $1000 in Joe’s account.  Jane would still have her $1000 checking account, rather than a bond. In both cases, the private sector’s assets went up by $1000, but now there is no federal debt to be repaid. The main difference in the private sector is that Jane now has a more liquid asset (checking account instead of a bond). Much depends on Jane’s propensity to spend that $1000 in her checking account. If Jane were inclined to just hold that $1000 in her checking account (as might be inferred from her willingness to buy a bond with it), then at least in the short term, the effect on the economy is about the same as if she had bought the bond.

On the other hand, if Jane spends that out into the economy, now her $1000 is added to Joe’s $1000 of spending. In an overheated economy, this dual spending would be inflationary.

If the economy is in deep recession, though, it might be helpful to have both Jane and Joe in a position to spend $1000. This stimulation effect would mean that the government could actually get by with spending less total money: e.g. instead of giving Joe $1000 and taking away Jane’s $1000 of spendable money, the government could give Joe only $500, but leave Jane with $1000 in her checking account by not issuing the bond. Then Joe’s $500, added to say $500 spending from Jane (if she spent about half of what was in her account) would give the desired $1000 of spending in the economy.

The circumstances of the government spending affect the benefits of issuing bonds versus printing money. One government priority should be to promote productive investments that will grow the overall economic pie. This purpose is well-served by giving the private sector sufficient financial assets, possibly including printed money, and letting private individuals and corporations allocate resources as they see fit. In recessions, heavy government spending is typically undertaken to inject additional spending into the economy. It may be beneficial to be able to accomplish this spending by issuing non-debt fiat money, as discussed above. Debt-free money could also finance a tax cut rather than increased government spending. This would also inject money into the economy, but let the people decide how that extra spending power is applied.

However, in normal (non-recessionary) times, the government is directly competing with the private sector for resources. For instance, government spends on its own salaries, and on programs like public education and infrastructure, which the private section would not necessarily take on.  To avoid high inflation, the government must divert financial as well as real resources from the private to the public sector. In wartime, the need of government to commandeer investment and production for its purposes becomes even more acute. This entails throttling down civilian production and consumption.

This can be accomplished in part by taxation, and in part by pulling money out of existing checking accounts by encouraging people to exchange this cash for government bonds. The First and Second World Wars were largely financed by war bonds. In our example above, this would correspond to Jane relinquishing her checking account money in exchange for a bond.

Printing New Money versus Retiring Debt

A proper application of this policy would involve measured steps, as with the Civil War greenbacks. Nearly every congressman longs to bring some federal spending home to his district, so there is a very real danger that the prospect of debt-free funding would unleash a gigantic and inflationary pork-fest unless Congress exercised notable fiscal discipline.

A less controversial use of debt-free funding would be to retire a certain amount of existing federal debt. Since a significant portion of the debt will never be repaid anyway (it will just get rolled over as yet more debt), this debt retirement seems like a relatively benign activity. If many U.S. government securities were retired by debt-free funding, it would make the remaining securities more desirable, and thus serve to hold interest rates down. Operationally, this could be done with the Treasury minting a trillion-dollar coin and giving it to the Fed in exchange for government debt, as discussed above. As hokey as it sounds, that would allow the Fed to maintain its double-entry bookkeeping, since it could carry said coin on its balance sheet as an asset, to replace the bonds it would have to relinquish. There may be other means of accomplishing the same thing purely on paper, without minting and securing a physical object.

While the initial purpose of U.S. securities is to finance the deficit, they also serve a useful function in the financial system as liquid, interest-bearing assets. Businesses, individuals, and pension funds need some ultra-safe investments. Also, having a reasonable market in U.S. securities can help support the function of the dollar as the world’s generic reserve currency, which has certain benefits for America.   These are reasons to take only measured steps in debt retirement, and not aim at eliminating all U.S. government securities.

In originally acquiring these government bonds from the open market during QE operations, the Fed had credited the reserve balances of the banks from which it bought the bonds. Thus, any bond-retirement operation between the government and the Fed would leave large amounts of reserves on the books of the commercial banks. This may complicate the unwinding of QE.

Currently, government securities must get transferred to the Fed for marketing, which is handled through the primary dealer banks. If the Fed wants to purchase government debt, it has to buy it back from the private banking system. The banks profit from all this sloshing back and forth. Before 1981, though, the Fed could buy government debt directly  from the Treasury. If debt-free funding is contemplated, it might be helpful to also reinstate the capability of direct purchase for part of the Treasure debt issuance. That would allow the debt retirement activities to operate largely independent of the Fed’s main open market operations that involve buying and selling securities to adjust the size of the reserves in the commercial banking system.

If the Fed simply bought the Treasury debt directly and held on to it, that would have much of the effect of debt-free funding. The Fed remits 95% of its profits to the Treasury, so if the Fed ends up holding many trillions of dollars of T-bonds, nearly all of the interest that Treasury pays on those bonds comes back to the Treasury. Paying interest on the national debt would become a trivial issue. At the end of 10 years, when a note became due for full repayment, the Fed could simply roll it over into a new 10 year note. Although the amount of debt held by the Fed would keep growing and growing, it would be so sequestered that it would be almost without effect.

Is Hyperinflation a Danger?

A common reaction to proposals of non-debt government funding is fear of runaway inflation, as exemplified by 1920’s Germany and today’s Zimbabwe.  Cullen Roche analyzed instances of hyperinflation in the past century, and pointed out that:

Hyperinflation… is a very different animal than inflation. Hyperinflation is a disorderly economic progression that leads to complete psychological rejection of the sovereign currency.

 Contrary to popular opinion, excessively high deficit spending and exorbitant government debt levels are not the actual cause of a hyperinflation. In most cases they have been the result of other exogenous events such as ceding of monetary sovereignty, war, rampant corruption or regime change. It is these exogenous events that result in the public’s rejection of the currency, a collapse in the tax system and the government response of printing more money to fill in the confidence void. Ultimately the confidence void cannot be filled and the currency is fully rejected by the public in the form of hyperinflation.

The German Weimar Republic was faced with an impossible situation after World War I: the victorious Allies demanded higher reparations (in gold) than could reasonably be paid, so the German government printed marks to exchange with foreign currency in order to buy gold. As the value of the mark fell, more and more had to be printed to continue the currency purchases.  When the French army occupied the industrial Ruhr area in order to collect reparations in the form of goods, the workers in the Ruhr went out on strike, and the German government printed even more money to pay them. By 1923 the mark had dropped in value by a factor of about a trillion, stressing society and ruining many German households.

In the case of Zimbabwe, the dictator Robert Mugabe took over the prime agricultural land from its mainly white owners and gave it to his agriculturally incompetent comrades, starting in 2000. Before then, Zimbabwe was a prospering net food exporter. After Mugabe drove out the former farm operators, internal food production declined. The change from being a food exporter to a food importer led to a shortage of foreign currency and difficulties in obtaining critical imports. These factors, combined with international sanctions, drought, and widespread HIV led to a general economic collapse. As is practically always the case, the hyperinflation in Zimbabwe was a symptom, not a cause of economic distress.

Provided the usual constitutional budgetary processes are followed, as they have been for over two centuries, there is no reason to fear that using debt-free funding would cause hyperinflation to suddenly erupt out of nowhere. The Congress, in collaboration with the President, determines what is to be spent each year. The U.S. government is going to spend what it is going to spend; the prospects of massive budget deficits do not seem to slow it down. Whether 100% of those deficits go on the books as the usual “forever debt” (which realistically will never be paid down, but will only continue to grow as it gets rolled over), or whether some of those deficits or debts are paid or retired by non-debt funding, is a rational choice which ought to be debated on its merits. If debt-free financing becomes a subject of national discourse, we may expect bankers and the recipients of their largesse to mount vigorous objections, since the commercial banks profit mightily from the current debt financing.

Is the National Debt Really a Problem?

The size of the on-going U.S. government deficits and debt is a subject of anxiety and debate. Pundits and politicians often make statements like, “You couldn’t run a business or household like that.” That is correct. The crucial difference is that a sovereign government has unlimited capability to create money, whereas a business or household does not. So these simplistic comparisons are inappropriate.

The current and historical figures for the U.S. public debt are available at the Treasury’s The Debt to the Penny website.  The  usdebtclock site displays all sorts of numbers (government debt, mortgage debt, student loans, trade deficit, Medicare liabilities, M2 money supply, GDP, etc.) updating in real time.

As of mid-February, 2013, the gross federal debt was about $ 16.5 trillion. This is just over 100% of GDP. However, about $ 4.8 trillion of that debt consisted of obligations of one sector of the government to another sector. While these intragovernmental holdings are counted within the debt ceiling, they are not considered as consequential. This leaves $11.7 trillion as the debt held by the public, i.e. by parties outside the federal government. About half of this public debt is held by foreign investors (e.g. the central banks of China and Japan), about 15% by the Fed (as a legacy of QE), and most of the remainder by pension funds, state and local governments, and insurance companies. It should be noted that American investors hold trillions of dollars of investments in the debt of other countries, so international debt holdings are not all one-sided.

It is not true that the U.S. needs China or other nations to buy its debt in order to fund the government. The primary dealer banks are required to make a market in Treasury securities, and the Fed could always pick up the slack in an emergency.  It serves Chinese currency and growth goals to return dollars to the U.S. as some sort of investment.

Prior to 2011, the revenues from Social Security taxes exceeded the payouts to retired citizens. The Treasury spent this extra money each year on other budgetary items, but compensated the Social Security program by issuing IOU’s.  These obligations comprise about $2.7 trillion of the intragovernmental holdings. This arrangement has artificially depressed the public debt figures. The only realistic means for the Treasury to make good on these Social Security Trust Fund obligations (in addition to paying for its regular budget items) in coming years is to borrow these funds from the public, this time explicitly.

Below is a historical chart of the debt held by the public as a percentage of GDP. This debt measure reached 73% in 2012. This is much higher than any time in the past except for World War II, but is comparable to the current public debt of other nations like Britain and France. However, it may go much, much higher in coming decades. The “Alternative Fiscal Scenario” is the projected trajectory if no major changes are made to government spending patterns. The GAO projected  that as the Baby Boomers retire and age, the Medicare/Medicaid/Social Security entitlement programs, plus interest payments, will balloon to the point of consuming 100% of government tax revenues shortly after 2030.

In the near term, there is no need to worry about the funding of the government debt, yet intuition suggests that this sort of future debt trajectory is not sustainable. However, it is not clear exactly how much debt is too much.

Padoan et al. believe that, below some threshold, public debt has little negative effect on economic growth.  However, their theoretical model indicates that at higher debt levels there is “the possibility of a vicious circle of high debt, low growth and unsustainable public debt dynamics.”  Their analysis indicates that “on average, a country recording a [public] debt ratio above 106% would see its debt ratio spiral out of control and its economy slump in the absence of offsetting policy action.” The U.S. might reach that point between 2020 and 2030, on its present course.

A 2010 empirical working paper,“Growth in a Time of Debt,” by Harvard economists Carmen Reinhart and Kenneth Rogoff has been the subject of intense controversy. This paper found that countries whose government debt/GDP ratios exceeding a threshold of 90% experienced a 1% lower growth rate in GDP.  As its findings were used by politicians to justify more austere budgets, this paper came under increasing scrutiny by other academics.

Robert Schiller in 2011 pointed out a problem with the implied causality in this paper: a correlation between high debt and slow growth did not necessarily mean the debt was causing the low growth. It was equally likely that the high debt was the result of some underlying problem which was also stunting growth.  In 2013, Herndon et al. at the University of Massachusetts reanalyzed Reinhard and Rogoff’s data set, finding significant conceptual and computational errors which discredit the original 90% threshold claim. Reinhard and Rogoff acknowledge those mistakes, and have backed away from positing a particular trigger level for debt problems.

In response to Krugman’s professional and personal attacks, Reinhard published an open letter rebuking his “spectacularly uncivil behavior.” Reinhard noted that she and Rogoff have not advocated for premature cuts in government spending; in fact, their work has taught policymakers that recovery from this sort of financial crisis is quite difficult, and demands  aggressive fiscal and monetary action. They call attention to the need for structural debt reduction in Europe. This is a more feasible operation (via ECB purchases of private as well as sovereign debt, following the lead of the Fed) than more Keynesian stimulus spending, which Germany would oppose and which the periphery cannot afford.

Historically, the public debt burden is managed not by paying it off, but by growing the economy faster than the debt. As noted earlier, fully paying off the national debt would tend to suck too much money out of the economy. (The exception to this is if a nation runs a persistent strong trade surplus, then extra money comes into the country in exchange for the goods flowing out of the country; this could allow the government to pay down its public debt without starving the private sector of financial assets.)

The dollar amount of U.S. government debt has grown steadily over the years, but post-war GDP growth has mitigated its effect.  If U.S. GDP growth in the next decade continues to be anemic, the debt burden will be felt more strongly. If natural, organic economic growth fails to appear, then increasing deficit spending may be called for to spur growth. However, that deficit spending will add to the debt, perhaps faster than the GDP will grow. Australian economist Richard Wood analyzed this problem and concluded:

There is a strong basis for considering deficit monetisation as the most likely optimal macroeconomic policy paradigm needed to simultaneously address both falling demand and rising public debt.  When policy interest rates approach their zero bound, deficit monetisation represents perhaps the only combination of monetary and fiscal policy instruments that could deliver a stimulus – that is, economic growth – without increasing public debt.

By “deficit monetization” Wood means the debt-free, fiat funding mechanisms discussed at length above.

With much excess capacity and many underemployed people, the current problem seems to be a lack of buying power rather than a lack of productive capacity. In Keynesian thinking, this type of problem is solvable by just throwing enough money at it, no matter how the money is spent. Keynes himself used the example of digging holes in the ground. Krugman has claimed that a massive deficit spending effort to counter an imagined threat of alien invasion would end the slump within 18 months, in the same way that (in his analysis) the enormous deficit outlays associated with America’s entry into WWII ended the Great Depression.  It seems clear, however, that if the government is going to spend extra money, society would be better served if that public money is invested in ways that create lasting benefit, such as infrastructure, subsidized solar panels or prenatal health care.  Governments have historically used public money to build infrastructure such as roads, parks, schools, and other public buildings that benefit citizens in general, but would be difficult to finance on a pay-for-use basis.

If another two trillion dollars in deficit spending would re-establish economic normalcy, such that deficits subside thereafter, that seems like a reasonable proposal.  Employing debt-free financing would offer a way to provide a large stimulus without more to the public debt burden. However, there is no guarantee that this added stimulus would really fix matters, as discussed below.

Past the 2030 timeframe, no amount of financial juggling can avoid the need to redefine the federal entitlement programs. There is too much intergenerational real resource transfer involved. The problem will likely shift from a lack of money to a lack of productive capacity (i.e. too few workers, and many of them engaged in personal healthcare tasks which are less amenable than manufacturing to big increases in productivity).

The Social Security setup basically has the current workers paying the benefits of current retirees. When the Social Security system was set up in 1935, the life expectancy of retirees was much shorter than it is now, and the ratio of workers to retirees was much higher. If the system continues to enable perfectly healthy 67-year-olds to retire, that places a greater burden on those who are working. Fewer people in the workforce tends to mean lower GDP; low real production means that boosting spending can only produce inflation. A crucial issue is whether on-going productivity gains will be large enough allow each worker to support a growing number of retirees. Printing money to pay the non-workers can grant them goods and services without increasing the national debt, but it would act as a further tax on the workers, since (due to inflation) their earned dollars will buy that much less goods and services.

It is estimated that Social Security outlays will rise modestly (as a percent of GDP) from about 5% now, to about 6% of GDP due to aging of the population as we approach mid-century. Health care costs are expected to be the real budget-buster.   Greenlaw et al. note, “Health-care-related expenditures are slated to explode if no policy action is taken. For example, CBO baseline estimates show federal government spending for Medicare and Medicaid rising from 5.5% of GDP in 2010 to more than 9% of GDP by 2030. The dramatic escalation in projected Medicare and Medicaid outlays is being driven by two factors — demographics and rising per patient expenses.” We will not review the details here, but the U.S. healthcare model is in need of drastic overhaul: The U.S. spends about twice per capita on health care as do other developed countries, with poorer results as measured by infant mortality and expected lifespan.

The Debt Ceiling and the Fiscal Cliff

The U.S debt ceiling is a limit set by Congress on how much debt can be issued by the Treasury. It was first introduced in 1917, and has routinely been raised by Congress as needed to allow the Treasury to fund the spending which has been authorized by Congress. For Congress to authorize spending, but deny the Treasury the means of funding that spending makes little sense. The normal time to limit spending is during the debate and passage of the bills that tell the government to do the things that cost the money.

Only in recent years has the debt ceiling been an instrument of political maneuver. In 2011 Republicans used it as leverage to obtain agreements for more fiscal restraint. A delay in raising the debt ceiling led to a stock market crash, a downgrade in Standard and Poor’s credit rating for America, and billions of dollars in increased borrowing costs. If the debt ceiling is not raised in time for the full funding of the budget, the Treasury can take “extraordinary measures” to juggle accounts for a couple of months, and then would do things like stop sending social security checks. It is unlikely that politicians would want to take the heat for that, but we can expect ongoing controversy.

The “fiscal cliff” refers to a set of tax increases and spending cuts that were due to take effect in early 2012, based on previous budget legislation. The projected decrease in the federal deficit was projected to push the economy into a mild recession. Most of the tax increases were negated by the American Taxpayer Relief Act of 2012, which was signed into law on January 2, 2013. The spending cuts, which would be apportioned by a “sequestration” procedure, were delayed until March 2013.  Mandated or entitlement programs (Social Security, Medicare, etc.) and debt interest comprise about 2/3 of the federal budget, so the budget cuts are focused on the remaining segments of the budget such agencies and civilian defense employees.

Japan: Pioneer of Astronomical Public Debt

Japan experienced a huge bubble in asset valuations in the late 1980’s, which burst in 1991-92, and then continued to drift downward. By 2000, stock prices had dropped by about 50%, and real estate prices by about 80%. Many real estate and other loans were made during and after the boom years became unpayable after the collapse in prices.

In an effort to mitigate deflation and restart the economy, the Japanese government has run an epic series of deficits for two decades. Current prime minister Shinzo Abe proposes aggressively increasing deficit spending, and has pressured the Japanese central bank into expanded QE, in an effort to increase the rate of inflation to a targeted 2% per year. The results of “Abenomics” are mixed so far. The Japanese stock market initially ramped up but now has largely retraced. The value of the yen has fallen, which should boost Japanese exports but has also made imports more expensive, so Japan continues to run a trade deficit.

Public debt is well over 200% of GDP and rising fast. Financiers have long predicted that buyers for that debt would start to balk, and that “bond vigilantes” will demand higher interest payments on Japanese bonds. Those who actually entered into this “widowmaker” trade against Japanese bonds have lost money, as interest rates on Japanese debt have remained close to 0%.  Domestic buyers have continued to purchase Japanese bonds, as has the central bank.

Japan can always create more yen to pay off its bonds. As long as excessive inflation does not take hold, those yen will still represent a more or less constant claim on Japanese real assets, so buyers are willing to keep buying yen-denominated bonds. So far, the high levels of debt have been sustainable. It remains to be seen what happens as the debt continues to climb and the Japanese population shrinks.

The past twenty years of Japanese stagnation are a cautionary example for other developed economies. Endless rounds of deficit spending and QE have been unable to re-ignite economic growth. Is this what the future holds for a post-2007-bubble America or U.K. or southern Europe? Keynesians claim that Japan could have righted itself if it had done even more deficit spending. Austrians might suggest that the deficit spending has actually prolonged the crisis, by enabling the Japanese government to prop up insolvent banks and businesses instead of letting them fail and clear the way for more viable enterprises.

Greece and Other Eurozone Countries

The woes of peripheral European countries are described here and here. Greek public debt is well above 100%. This debt is owed in an external currency (the euro). Creation of euros is in the hands of the European Central Bank (ECB).  The only way to pay back this debt is for Greece to run a trade surplus, especially with its euro-using neighbors. However, Greek labor costs (factoring in relatively low productivity) remain high compared to other European countries. There are historic structural factors that also hinder Greek finances such as corruption, a bloated public sector and tax cheating.  Austerity measures designed to reduce the government deficit have decimated the economy, making even harder for Greece to service its debt. With no domestic job prospects, educated young Greeks are fleeing the country, which further hinders Greek economic growth.

A common worry is that, with its ever-growing deficit, America will “become another Greece.”  There is no chance of that happening. If Greece still had its own drachma currency, the Greek government and central bank could create more money as needed. If Greek finances deteriorated, the value of the drachma would drop relative to other currencies. This devaluation would automatically make Greek labor cheaper to the rest of the world, and improve their trade balance.  But none of these options are available to Greece, since it uses the euro.  The position of Greece or Spain is like that of an individual American state, or a household or business. Greece is a user of an external currency. This is true for all 27 countries that use the euro.  On the other hand, America can issue currency as needed, so it can never run out of money. The U.S. may run into problems with inflation and stagnation, but never with solvency.

 Goverment Fiscal Policy: Path to Prosperity or Road to Serfdom?

The financial operations described above enable ongoing government deficit spending. Some summary points are:

–  Persistent government deficit spending can be a responsible policy.

–  Running a large trade deficit practically guarantees a large government deficit, although as noted by Buchanan et al. the causation can run either way.  Cutting the trade deficit would necessarily improve domestic employment and domestic government deficits.  How to accomplish a cut in the trade deficit is not so clear; it will likely take grit and vision, and willingness to stand up against  against U.S. business interests that profit from foreign manufacturing.

– A government that issues its own currency  can always come up with more money to fund its deficit. Thus, the U.S. government is not subject to the same constraints as a private household or business.  The real near-term limit on the level of deficit spending is the need to avoid excessive inflation.

Deficit spending is currently funded entirely via issuing debt instruments such as notes and bonds. It is not clear whether future GDP growth will be sufficient to mitigate the increasing dollar amount of the federal debt. As with the Civil War greenbacks, partial funding by debt-free issuance (i.e. outright money-printing) is a policy which deserves consideration, particularly for retiring older debt.

Whether expanded deficit spending actually increases overall economic activity is not clear.  In normal times, the extra dollar spent by the government can simply crowd out or shut down a dollar of private spending, e.g. by leading to a rise in interest rates or price levels. A 2011 IMF study by Ilzet, et al. concluded that the net effect of government stimulus spending for a countries like the U.S. with high debt levels and floating exchange rate is typically zero or even negative.

In near-recessionary times like these, crowding out is less likely. When aggregate demand is muted despite near-zero interest rates, Michael Woodford’s analysis predicts a multiplier (i.e. dollar net increase in GDP per dollar of stimulus) of at least one. Estimates of the effects of the 2009 stimulus package vary widely, though most economists seem to believe it had some net positive effect.

The quality and long-term effects of that stimulated activity are less clear. It may simply pull future consumption into the present.  How to promote growth which can sustain itself after the stimulus of deficit spending is removed remains an active area of research and debate in the field of economics. It seems self-evident that paying people to produce some asset of enduring value (e.g. renewed transportation infrastructure) is a better use of government funds than paying people to sit on their couches.

Some schools of economics are generally averse to government “interference” in the economy, claiming instead the the business cycle of easy credit/over-investment/boom/crash/credit contraction just has to work itself out, no matter the short-term cost. However,  in our present system the government spending and taxing policies do inevitably result in some level of net injection of money into the private economy, so it is impossible for the government to not “interfere” in the macro economy.

I can see the virtue of letting inefficient individual companies fail quickly and completely, to release resources for more effective enterprises.  But there are economic and societal costs to otherwise sound businesses failing solely  due to recession. If Amy has to close her flower shop in the trough of a one-year recession, leading to her losing her home and other complications, and then Kathy comes in and restarts essentially the same business in the same shop, that seems like a preventable tragedy rather than beneficial “creative destruction.” It’s not clear that forcing citizens to stew for years and years in underemployment serves to “clear markets.”  Thus,  a certain amount of counter-cyclical, safety-net government spending seems justifiable.

That said, the manner in which the government injects money into the private economy has implications for the freedom of its citizens.   Attempts by government planners to manage investment and spending often result in unproductive pork, or worse:  To inject money by  government spending programs inherently involves  favoring some people and organizations over others. This can become a means for the incumbent officials to enforce their prejudices and prolong their reigns.   Vigilance will always be needed to guard against this tendency for incumbents to harass their opponents and favor their supporters by means of government contracts and regulations.

Also, there is evidence that socially attractive activities such as providing for the welfare of the poor are more effectively managed by local efforts than by national programs.  Some forms of government transfer payments tend to build long-term dependency, perversely reducing the incentives for individuals to re-enter the productive workforce.

If it is true that government deficit spending is required to provide net financial assets to the private section, we suggest doing this in a neutral or blind fashion, e.g. by uniform tax cuts (same dollar amount for every wage-earner, not to exceed his or her tax liability). This would minimize government intrusion, and allow the people to make their own decisions on how to handle their own money.

February 2020: Epilogue

Some things have changed and some things have remained the same since I wrote this article in 2013. Everything said about how the monetary system operates remains valid, as far as I know. What I did not foresee was that central banks around the world would continue to intervene with purchasing trillions of dollars’ worth of bonds and related assets, and keeping short term interest rates near or even below zero for years and years.  It is interesting today to look back to all the macroeconomic uncertainties seven years ago, and see how things have actually played out.

In 2013, the world was in the midst of a painfully slow recovery from the traumatic 2008-2009 Great Recession. This recovery was much more sluggish than the recoveries from all other post World War II recessions, where lost jobs were nearly all recovered within about 2.5 years. Central banks were taking unprecedented measures to try to stabilize finances and help economies regain growth.

Europe was still in the throes of a sovereign debt crisis. As of 2013, Greek public finances were still shaky, with fears of default driving 10-year government bonds yields over 10% and Greek populist political parties fighting the austerity measures demanded by the EU as the price for a bailout. At that point, the fiscally-responsible Germans were preventing the European Central Bank (ECB) from simply underwriting the bonds issued by European countries like Greece. However, since then the ECB has devised ways to do stealth quantitative easing – – it provides money at very low interest rates to national banks to enable those banks to purchase government bonds. The ECB balance sheet now stands at over 4.5 trillion euros. This has driven government bond interest rates down and down, to zero and sometimes even negative. Because of the implicit ECB backstop, Greek bonds were recently issued with an interest rate of under 1%, which is lower than the rates for current US Government debt. No one could have imagined such a thing in 2013. Such is the power of central banks.
In Japan, the central bank has gone wild, buying anything and everything including stocks as well as bonds. The Japanese central bank routinely vacuums up a large fraction of the debt issued by the Japanese government. This has kept a Japanese interest rates near zero for decades. For some years, various finance commentators have warned that private bond buyers would rebel against the huge government debt and deficit spending, and would force Japanese rates back up. But all the experience of the last 10 years illustrates the truth that you cannot fight the central banks. They have for all practical purposes, infinite resources to execute their objectives, since they can create as much money as they want with the tap of a keyboard.

The Swiss national bank (SNB) currently engages in raw currency manipulation by creating billions of Swiss francs out of thin air, and using them to buy things like Euros and U.S. stocks. Dumping all these francs into foreign markets serves to reduce the foreign exchange value of the franc. This is done in order to make Swiss exports more competitive. A side effect of this currency war operation is that the SNB has come to own huge stakes in U.S. corporations, e.g. it owns more publically-traded shares of Facebook than Mark Zuckerberg.

In the U.S., the Fed intervened massively to stabilize the financial markets in the wake of the 2008-2009 meltdown, but those interventions were seen as extraordinary. But the extraordinary has since become the ordinary. The Fed balance sheet remains bloated at over $4 trillion, and there is no realistic prospect that it will ever shrink down low enough that the Fed can resume setting short term rates by open market operations, like it did before 2008.

In the U.S., the Fed under Chairman Powell tried to reduce its holdings and to increase (“normalize”) short term rates back up to 3-3.5% or so in 2018. That ended disastrously, with a stock market meltdown at the end of the year, and with the Fed quickly back-pedaling and ratcheting rates back down in 2019. In autumn of 2019, the commercial banks got overwhelmed with the flood of bills issued by the U.S. Treasury due to the ballooning federal deficit, and the “repo” market (which I will not try to explain here) froze up, so the Fed intervened by buying lots of Treasuries, in the process re-inflating its balance sheet.
The bottom line is that central banks have succeeded in enforcing extremely low interest rates over the past decade, and at this point it looks like low rates are here to stay. Private and public debt has ballooned to such enormous amounts, that any large increase in rates would just probably crash the whole system, since so many parties simply could not keep going if they had to pay higher rates on their debt.

Here is a graph of nominal bond rates over the last 700 years:

Source: Article by author BDCBuzz, https://seekingalpha.com/article/4322926-dividend-stocks-are-new-bonds-for-retirement-portfolios , who credited it to VisualCapitalist.

The long term trend has been toward 0-2% nominal interest rates, or essentially 0% or negative real rates (i.e. after inflation). Billions of dollars’ worth of sovereign bonds now trade at negative nominal rates; you would do better stashing your money under your mattress. Corporate and other bonds of course trade at somewhat higher rates than this government debt. This is a novel global macroeconomic experiment, being run in real time. One wonders how economists a hundred years from now will assess it.

So, what are some of the effects of more or less permanent near-zero interest rates? One effect is to enable massive government deficit spending, in Japan and now in the US. It seems like neither major American political party cares about federal deficits any longer. One party used to at least posture about being concerned about it, but their reward was to be bashed as “heartless” by the other party for reining in domestic spending, so they seem to have given up that fight. Politicians now seem to assume that the Fed will step and buy government bonds (“monetize the debt”) as needed to keep interest rates down. It seems that in practice this plank of “Modern Monetary Theory” has become the new orthodoxy.

It used to be the case that the Fed was so independent that it would enforce financial discipline on federal government. If the federal deficit spending started getting high, with attendant inflation, the Fed would crank up interest rates to cool down the economy. This would (in theory, at least) punish the profligate president and Congress by essentially causing a recession on their watch, with all the human pain of job losses. If this punishment caused the stock market to crash, so be it. But now the Fed seems to be more or less hostage to the stock market. If the market swoons, the Fed has shown that it will quickly lower interest rates.

If Europe, the ECB sits above any one country, and is heavily influenced by the Germans, with their tradition of strict financial discipline. Although the ECB has provided cheap money to keep interest rates low in European countries, it demands in return from those countries that they meet certain austerity targets. This has to some extent forced those nations to keep their deficit spending under control. On the one hand, that may seem admirable, but in practice, European economic growth has lagged far behind that of the US in the past decade, perhaps because of all the austerity.

In the U.S., the more productive coastal states subsidize the poorer interior/southern regions via federal transfer payments such as Medicare and food stamps. This money recycling helps allow the folks in the poorer regions to keep purchasing goods and services produced by the wealthier regions, so it is a win-win. However, in Europe, the more-productive Germans and Dutch are reluctant to simply ship their hard-earned euros to Greece and Spain to subsidize the lifestyles there.

In the U.S., companies have taken advantage of low interest rates to issue staggering amounts of corporate debt. Rather than using this money to invest in new factories and hire more workers, a large fraction of it has gone into simply re-purchasing corporate shares. This in turn has driven U.S. stock prices higher and higher and higher. For a given company, the underlying business may not change, i.e. the total revenues and earnings may not grow much, but because fewer shares are in circulation, the earnings per-share continue to creep up. For such brilliant financial engineering, management rewards itself with fat compensation packages.

Looking at things from a more macroeconomic perspective, it seems that permanently low interest rates have a paradoxical “zombification” effect. Japan is a pioneering example of this, since low interest rates and slow economic growth have been a feature there for several decades now. Apparently, the low interest rates enable poorly-executing, walking-dead “zombie” companies to stay in business, whereas previously they would have gone bankrupt and disappeared. On the bright side, this tends to keep employment relatively high, which has been a general feature of many developed economies over the past decade. But on the other hand, this inhibits the “creative destruction” process whereby more productive companies replace less productive enterprises. So the real per capita economic growth remains subdued. Also, large companies seem able to better take advantage of low rates, so we see big companies getting bigger and more monopolistic, to the detriment of smaller firms which might otherwise be introducing new ways of doing things.

A zombified economy where nearly anyone who wants to work can get a job, even at low and stagnant real wages, seems like a fairly benign outcome. However, if young people see only a terminally dull future ahead for themselves, this may lead to dissatisfaction which could have political consequences. Moreover, if the real interest rate of investment grade bonds remains zero or even negative, that will challenge pension funds in meeting their commitments to pay pensioners in coming decades (many pension funds still assume they can obtain secure long-term returns of around 7% on their portfolios). It also calls into question the traditional 60/40 stock/bond portfolio for individual investors, if the 40% bond portion is earning practically nothing. This impelled me to look for, and actually find, some ways to earn reasonably high (6-8%) yields in the present environment (see High Yield Investments).
Every age has its reasons for worry. As I post this, the Wuhan corona virus from China is spreading throughout the world. Hundreds of thousands are sick, and thousands have died. Whole cities have been put under quarantine. Here in the U.S., surgical masks (which are all made in China these days) are all sold out, and organizations are making contingency plans for scenarios where employees, teachers, and students are all confined to their homes for indefinite periods. The upcoming election may pit an erratic incumbent against a socialist. Interesting times – – we will see if the Fed can keep working its magic.