Living long enough to collect

The Social Security system and the Old Age, Survivors, and Disability Insurance program (OASDI) were created by the Social Security Act of 1935.  Social Security is the key part of a set of public social insurance programs in the United States because of its near-universal coverage of Federal and private sector employees.  The program is immensely popular with the American public, but its size resulting from the scope of its coverage also makes it a symbol of excessive government to that portion of the public that wishes to shrink the central government.  It is politically difficult to attack the program outright, even though the program might benefit from some reform.  A different strategy is to reduce Americans’ dependence on the program by keeping the benefit payments small.  This has been justified by claiming that when the program was started, life expectancy was equal to the retirement age; the average contributor wasn’t expected to live long enough to collect.  The data presented here argue that this claim is false.  In examining this claim, it may be possible to address some other issues in actuarial estimates of life expectancy and see what factors are impacting the age composition of American society.

At any moment, each person is faced with a probability of dying.  The probability varies across one’s life, but toward one’s later years its central tendency is to increase; death is certain at some point.  For any individual, the exact probability is likely unknowable.  Actuaries have constructed functions or life tables that approximate for a model population, the probability of death for each surviving individual within the year.  The model is a count of the number surviving out of a population of, say, one hundred thousand individuals starting at the same time.  The reduction in the number of survivors by the end of any year divided by the number at the beginning of the year estimates the probability of death within that year.   Life tables have been produced since the 17th century.  The rates of mortality will vary with the population and its age as advances in medicine and public health change the odds.

For the purpose of this narrow question, however, let us use the Commissioners 1941 Standard Ordinary Mortality Table from the Actuarial Society of America because it is close to the data available to Congress when the Social Security Act became law.  The table is based on the experience of white males.  A table for white females would have longer lives and tables for non-whites would have shorter ones.  This table was selected because it would be the model of who was in the workforce in 1935.  (This table is also attached below.)  The table lists five variables:

  • x = age at beginning of the year
  • l(x) = the number living at the beginning of the year
  • d(x) = the number dying during the year
  • p(x) = the probability of surviving the year, or (1 – d(x)/l(x))
  • e(x) = the number of years left that an individual of age x can expect to live.

Manipulation of these values in the table allows one to calculate such things as the chance of dying between any two ages or similar questions.

Consider some key points:

  • At age 0, e(x) is 62.3 years.  That means life expectancy at birth in this table is about 62 years and 4 months.  That would seem to match the claim that Social Security was designed so that workers could not expect to collect the benefits .  Note, however, that p(x) is 97.74% which means that newborns had about a 2% chance of dying within that first year.  As will be discussed later, childhood mortality exerts a major impact on life expectancy at birth.  By the same table, those aged 5 will number 983,817 out of 1,023,102 born five years earlier.  This suggest a 3.8% chance of dying within the first five years for this population.  It should be noted that for those who survive to age 5, total life expectancy is 59.8 additional years for a total life of 64.8 years.
  • Newborns do not work.  Given child labor laws, even in 1941, one did not expect to work before age 16 at a minimum.  The population that reached at least 16 is 960,201 out of the original population for a 93.8% survival rate from birth.  At this point, many hazards have been met.  The remaining life expectancy at age 16 is 50.1 years for a total life expectation from the start of participation in the  workforce of 66.1 years.  At this point, the claim loses validity: anyone entering the workforce has, on average, an expectation of living 13 months into retirement.
  • At age 40, one is about halfway through one’s working years as that is halfway between entering the labor force at 16 and retiring at 65.  Although the risk of mortality within the year is slightly greater than it was at age 16, it has increased at a slower pace than life expectancy.  If one survived to 40, one had the expectation of living to 69.3.   At that point, 25 years before retiring, one could expect to enjoy Social Security benefits for at least four years.
  • At age 65, upon retirement, e(x) is 11.6 years.  That is, total life expectancy for new retirees would be 76 years and 6 months.  The total population that started working at 16 has been reduced by almost 40%, but more than 60% of that group will reach retirement age.   But half of that loss has occurred since age 53, or within a dozen years of retirement.  Clearly, the claim that Social Security was designed such that retirees were never intended to collect the benefits (because life expectancy was less than 65 years) is absolutely false.

The underlying issue this exercise points to is how to interpret life expectancy and interpret what affects its value.  Looking at single values such as expectation at a specific age does not give sufficient information to draw conclusions about the dynamics involved.  Consider four life tables: English Life Tables No. 8 for Males 1910-1912; Commissioners 1941 Standard Ordinary Mortality Table; Commissioners Standard Ordinary (1958 CSO) Mortality Table for Male Lives; and the Social Security mortality table for males of 2010.  There are some methodological differences among them and they address somewhat different populations.  Nonetheless, they tell a consistent story about changes in life expectancy across the past century.

Across these four tables, life expectancy at birth was 51.50 years, 62.33 years, 67.55 years, and 75.96 years respectively.  This would indicate that across the past century, an expected male lifespan had increased by approximately 24 years!  Does this represent advances in gerontology that increase the chances of living well into old age?  Look instead at the life expectancy of someone attaining age 21.  For the four tables the values are 64.37 years, 66.66 years, 70.33 years, and 76.96 years.  The spread in lifespan across the tables has decreased from 24 years to 12 years.  At age 65, the values are 75.99 years, 76.55 years, 77.97 years, and 82.61 years.  The range has dropped to seven years, but only the 2010 table shows a major improvement in life span.

What has happened is seen in graphs that compare the annual mortality probabilities and the life expectancy by age among the four tables as shown in the two attached graphs.  The key problem is that is one does not have a good chance of reaching 5 years or 16 years or 21 years, then one has a much lower probability of reaching old age; life expectancy at birth is lower.  Across the four tables, the greatest spread of estimated lifespan occurs at birth.  The English curve shows a steep climb during the first year.  That is, life expectancy climbs very quickly during that year for those who survive it.  What is at play is a significant reduction in childhood mortality and a general improvement in survival probabilities for the rest of one’s life.

This is more clearly seen in the second graph, Annual Mortality Probability.  The lower a curve sits in the graphed space, the lower is the probability of death at any year.  Mortality tables from later years – 1941, 1959, and 2010 – show a lower probability of death at every age.  The major area of improvement in terms of lower mortality rates is particularly in the area between birth and age 20.  In general, the more recent the life table, the lower the mortality rates.  This even holds up to age 40.  Beyond that, three of the curves lie fairly close together; there was only marginal improvements in mortality over the years from 1910 to 1960.

Note that for the English Males data, mortality in the first year of life is approximately 12%.  Approximately one in eight infant boys would not live through their first year.  In calculating a weighted average – life expectancy is just that: an average of ages weighted by the share of the population that achieves it – adding a large zero for twelve percent of the population will pull down the average.   This explains why the spread among the four estimates of life expectancy at birth was so wide and then narrowed for later ages.  For males who survived to age 20, the probabilities of death were fairly similar for each year thereafter under three of the tables.  Their expected life span from that point forward would be about the same.  Only the 2010 model shows a consistently lower mortality rate for all ages.  This probably reflects a general improvement in health across the whole of life, not just the improvement of childhood experience.

In summary, using life expectancy as a measure of life quality or such requires care in stating the proposition.  We have shown how it was misapplied in the case of Social Security.  A review of these life tables then highlights what has been the real impact of better medicine and healthier lives over the past 100 years.  The improvements have had their greatest impact in childhood mortality.  If we have more old people alive in our society today, the reason may be less one of over-investment in gerontology as much as success in childhood health.


Table: 1941 Commissioners Standard Ordinary Life Table

1941 CSO

Graph: expected total lifespan by age attained

Expected Lifespan

Graph: annual mortality probability

Annual Mortality Probability



Realism in the politics of budgets

[This post is slightly different from those previous, including unpublished drafts.  Most posts have been based on looking at historical data to understand how we have met some of today’s issues in the past.  This post, however, is based solely on my own opinions derived from observations and work, especially in emerging market economies.]

Whenever I am asked to present an overview of U.S. fiscal operations to foreign delegations, if there is sufficient time, I devote part of my talk to addressing key institutional differences between the U.S. and most other countries.  I do this to reduce the risk that my audience might misinterpret the U.S. policies or outcomes because the listeners are building on a vision of their own domestic institutions.  The misunderstandings that arise from thinking all governments are the same can show up in the questions that cannot be answered as they are posed.  In particular, the differences between the logic behind budget formulation and execution, central bank policies, audit, and treasury management under what I suggest are two systems can impede a clear understanding of how each government has incorporated the principles of government finance into its framework.

I know of this problem at first hand.  My good colleague Stephen Bartos (then, a senior manager in the Australian Department of Finance and Administration) and I spoke past each other for quite some time trying to explain to each other how our treasuries functioned.  We could not see, for example, why the other claimed ministries [should/should NOT] be free to invest cash balances to augment their budgets.   Finally, we realized it was a difference in institutions, not a principles, that affected our views.  I credit Stephen Bartos for the label “Washington / Westminster split” that sums up the matter succinctly.  The essential distinctions the split refers to are these: how coherent are the political aims of the executive and legislative branches under either system and how does that coherence affect budgets and fiscal operations.

In a parliament – the Westminster system – the executive branch, that is, the prime minister and his cabinet, is formed from the majority party or the majority coalition that holds the legislature.  In this situation, there is every political incentive for the legislative party to want the executive team to succeed in executing the budget approved by the parliament.  Budgets are important because the choices they make express the aims of the majority faction to best satisfy the wishes of its supporters.  The logical extension of this thinking is that, as that faction holds the majority of the parliament, its wishes must reflect its supporters who are also the majority of the electorate.  (Nb: I am not asserting that this last view is necessarily true; I am claiming that it could appear this way to the faction in power.)  The budget becomes the realization of the party’s manifesto, the implementation of cabinet policy.  In practical application, this means that the line ministries can expect more leeway in executing the budget.  They may expect not to be held strictly to the limits set in the budget law, i.e., budget overruns have a good chance of being excused.  Further, the parliament is likely to accept additional expenditures as they are identified during the budget year.

The bigger problem, however, lies in budget formulation.  A state budget has three areas: proposed expenditures, expected revenues, and estimated gap financing.  Often, there is an established limit on how much debt may be used to fund the budget; the limit may be set as a percentage of the total budget or of GDP or as a numeric value in local currency units.  In an ideal situation, the gap financing and a realistic estimate of revenues should set the maximum value for budgetary expenditures.  The ideal situation rarely obtains; as the expression of the ruling coalition’s manifesto, the expenditure program is not to be left as a residual.  What appears to be the more frequent case is that the expenditure level is set and revenue is expected to cover the difference between it and the limit on gap financing.  Given the difficulties of forecasting revenue, there is a reasonable temptation to assume the best regarding receipts.  Sadly, experience has shown how inadequate this approach is.  Government revenues in many emerging market countries are imaginatively overstated.

The frequent result is the need to revise the budget at some point late in the fiscal year.  The budget can be expected to show a larger deficit than envisioned at the start of the fiscal year.  If expenditures must be cut to maintain balance, these will often be made on what remains.  This may not represent the budget choices that would have been made at the start of the fiscal year if revenues were accurately forecast.

In a republic with separate legislative and executive elections – the Washington system – there is no necessary political coherence between the party controlling the legislature and the party holding the executive branch.  One political party may hold the legislature while another may take the executive branch.  In the U.S., it is not uncommon for the government to be so split with the Presidency and the Congress in different hands.  In this situation, the power to set budgets becomes a means to exercise control over the opposing party in the executive branch.    The legislative branch jealously guards its power over the purse.  The limits set in the budget are held as fixed.  One consequence is that all funds are given to the treasury to manage.  No agency, with rare exceptions, is allowed to hold funds outside of the treasury lest they invest such funds in a way to augment the resources available to them.  Further, the legislative branch expects that the limits it establishes for agency spending will be strictly followed.  Agencies may spend no more than is authorized and may not spend any more than is authorized for any particularly purpose.  To enforce its discipline, the Congress has created the Anti-Deficiency Act.

The Anti-Deficiency Act (31 U.S.C. § 1341) prohibits federal employees from

  • making or authorizing an expenditure from, or creating or authorizing an obligation under, any appropriation or fund in excess of the amount available in the appropriation or fund unless authorized by law.
  • involving the government in any obligation to pay money before funds have been appropriated for that purpose, unless otherwise allowed by law.
  • accepting voluntary services for the United States, or employing personal services not authorized by law, except in cases of emergency involving the safety of human life or the protection of property.
  • making obligations or expenditures in excess of an apportionment or reapportionment, or in excess of the amount permitted by agency regulations.
Federal employees who violate the Antideficiency Act are subject to two types of sanctions: administrative and penal.  Employees may be subject to appropriate administrative discipline including, when circumstances warrant, suspension from duty without pay or removal from office.  In addition, employees may also be subject to fines, imprisonment, or both.
The consequence of this rigidity is that opportunities for economies or efficiencies may be missed when they present themselves for brief moments.

What does this mean for advisors who advocate rational budgetary processes?  When we argue for idealistic budget processes that look first to expected revenues and tailor expenditures to match available resources, we are advocating policies that are unrealistic.  They contradict the rational behavior of actors under the prevailing rules in their governments.  The expected outcome is a continuation of the budgetary deficits and lost opportunities.  The guidelines for ideal fiscal management that we are advocating violate the rationality of political agents and have little chance of traction.


About those debt limit “accounting tools.”

Whenever the United States’ government approaches its statutory debt limit, many people are confused by the ability of the government to continue operating for weeks or months beyond the initial announcement that the debt limit has been reached. The unfortunate side effect is that some begin to suspect it is all a fiction or is evidence of fraud. The continuance of operations is usually explained by reference to “accounting tools.” Recognizing that that explanation is not particularly clear, I will attempt to be more complete.

The Statutory Debt Limit (currently at $18.113 trillion) is the maximum amount of money the government is allowed to borrow without requesting additional authority from Congress. It is a self-imposed limit, not set by markets or creditors. The categories of debt which are counted in the limit include the $13 trillion of debt which was issued into the markets – bills, bonds, notes, and TIPS – to finance public expenditures AND approximately $5 trillion of debt that is held within the government or the public as the backing for the Social Security and other trust funds and a small amount for savings bonds. These are debt instruments which cannot be traded in the market and are called nonmarketable debt.  (See attached MSPD, p. 1.)

A word is needed to explain the trust funds. Just like private insurance companies or pension plans, when cash or premiums are paid in, the proceeds are not just kept in the vault; they are invested in safe instruments. That is the only way money can be earned to pay off future benefits. With funds like Social Security, at $2.8 trillion, it is held to be more prudent to invest them in government debt rather than in the equities markets.  (Listed in MSPD under Government Account Series – Intragovernmental Holdings, as Federal Old-Age And Survivors Insurance Trust Fund, p. 12.)  The Social Security Trustees buy a special series of Treasury debt at yields that are set to reflect the broader market rates over time. The cash can then be used by the Treasury to pay for current operations. This helps explain the paradox that gets quoted in online comments that allege that there was no budgetary surplus around the year 2000 because the debt subject to the limit still increased. Unless marketable debt is being actively paid down or bought back (that is, it is held steady) the flow of Social Security payroll taxes will continue to come in and additional trust fund debt will be issued to absorb them. Since both types of debt count in the limit, the debt goes up even in a surplus.

Three funds are important in the case of a debt limit contingency: the Exchange Stabilization Fund, the Civil Service Retirement and Disability Fund, and the Thrift Savings Plan Government (or “G”) Fund.

The Exchange Stabilization Fund (ESF) consists of three types of assets: U.S. dollars, foreign currencies, and Special Drawing Rights (SDRs), which is an international reserve asset created by the International Monetary Fund.  The ESF can be used to purchase or sell foreign currencies, to hold U.S. foreign exchange and Special Drawing Rights (SDR) assets, and to provide financing to foreign governments.  All operations of the ESF require the explicit authorization of the Secretary of the Treasury.  By law, the Secretary has considerable discretion in the use of ESF resources.  The legal basis of the ESF is the Gold Reserve Act of 1934. The dollar assets of the ESF, about $22 billion, are included in the debt subject to limit because they are held in nonmarketable debt that is reinvested overnight each day.  (See Exchange Stabilization Fund, Office Of The Secretary, Treasury, under Government Account Series – Intragovernmental Holdings, p. 12.)

The Civil Service Retirement and Disability Fund (CSRDF), under the management of the Office of Personnel Management, receives the contributions of Federal employees covered by the former civil service pension system that was closed to new entrants in 1987. The fund also invests its assets (about $700 billion) in nonmarketable debt issued by the Treasury. Each month, some CSRDF holdings (about $6 billion) are cashed out (disinvested) in order to pay civil service pensions and some funds are received as employee contributions.  (See Civil Service Retirement And Disability Fund, Office Of Personnel Management, under Government Account Series – Intragovernmental Holdings, p. 11.)

The Thrift Savings Plan (TSP) is a deferred compensation plan, like a 401(k), for Federal employees. They may choose from several funds issued by the TSP Board to diversify their savings among common stocks, corporate bonds, international stocks, and government debt (the G fund.) The TSP G fund is important because it is also held as Treasury debt subject to the limit and the trust fund assets ($191 billion) are reinvested overnight, each day.  (See Thrift Savings Fund, Federal Retirement Thrift Investment Board, under Government Account Series – Held by the Public, p. 11.)

Each of these tools are used to provide breathing room under the debt limit.

When the total of marketable debt and nonmarketable debt approaches the legal limit, the Treasury disinvests from these three funds to reduce the amount of nonmarketable debt. (There is really no new cash created, but the disinvested amount can be recorded as cash in the accounts.) What does happen is that the amount of nonmarketable debt thus cashed out can be replaced by an equivalent amount of marketable debt that does bring in cash. By replacing nonmarketable debt with marketable bills, notes, and bonds, the government can obtain cash to sustain operations while staying within the statutory debt limit.

For the TSP G Fund and the ESF, the operation is simple; their assets which are held as overnight investments are just not invested one morning. There is a corresponding drop in the national debt to match the amount which will not be invested overnight further. That same day, the Treasury can issue a matching amount of bills, notes, and bonds and receive the cash for them.  It is equivalent to actually getting the cash from the ESF or G Fund.  The funds are not completely liquidated at once.  Instead, the Treasury can disinvest bit by bit for the amounts it needs to finance government each week. The process can continue for as long as the combined assets of the two funds – about $214 billion in January 2015 – can support government operations.

The CSRDF is managed under different rules. The Secretary of the Treasury has the authority to declare that the investment of proceeds of the fund would create enough debt to push the government over its debt limit. The Secretary, therefore, declares a debt suspension period (DSP) – usually in months – during which he will not invest in the CSRDF. Further, as funds will have to be cashed out in order to pay pensions, the declaration of a debt suspension period will yield $6 billion for each month of the declared DSP.  In declaring a DSP, the Treasury can immediately disinvest about $6 billion for each month of the DSP.  The government, however, can still count on having to spend an additional $6 billion each month to actually pay its obligations. This was money that came from the CSRDF before the debt suspension and which now must be paid from general funds. The length of the DSP is not arbitrary; the Secretary could not, for example, declare a 10-year DSP which would give him access to the entire $845 billion in the CSRDF. The length of the DSP must be defensible and relate to a realistic estimate of how long, in months, it may take to get out of the debt limit stringency. (The longest in my memory was 18 months. Treasury lawyers were never overly easy on this point.) So a declaration of a ten-month DSP would add about $60 billion initially to the Treasury’s room under the debt ceiling.

All of these funds are made whole when the debt limit is increased.  The G Fund and the CSRDF must also be paid the interest they lost during their use as debt tools.

How long the resources available in these tools can be used by the Treasury to continue under the debt ceiling depends upon the time of year, (for example, are major tax collections due to supplement government coffers) and the rate at which expenditures are being made.   (These resources can all be seen listed in the Daily Treasury Statement.)  It is this variability from debt limit episode to debt limit episode as well as the subtle way of using unseen debt that can leave the citizen puzzled about what a debt limit crisis really means.


Table: Monthly Statement of Public Debt (MSPD), January 2015

Statement of Public Debt, Jan. 2015

Table: Daily Treasury Statement (DTS), January 31, 2015

Statement of Public Debt, Jan. 2015

Getting out of debt


RevIssue imagesO3361UJA

During the long wait for recovery after the 2008 financial crisis, a populist surge in American politics came to worry about the national debt.  It was pointed out that the US debt-to-GDP ratio was about unity: that we owed as much as we earned as a nation during a year.  The popular charge was that we were bankrupt.  Worse, it was claimed, we could never pay off this debt and that we were burdening the next generation.

(In those arguments must reside a good number of discussion topics.)

The important point is that, in addition to once paying off our entire national debt, we have made serious attempts at debt reduction.  Not all of the methods are healthy for the economy.

During the United States’ civil war, the costs of restoring the Union drove the national debt from $64.8 million on July 1, 1860 to $2,680.7 million on July 1, 1865 or more than 41 times over and equal to eight times the government’s revenue.   Yet, by July 1, 1890, twenty-five years later, the national debt stood at $1,552.1 million; it had been reduced by more than 40% from its high value.  It was done by the stamps, and other measures, you see here.  The U.S. initiated a series of excise taxes that caused a budget surplus and allowed the paying down of the debt. Rev Stamp playing cards

To finance its expenditures during the war, the United States engaged in new means of finance.  Prior to 1860, customs duties, (at $39.6 million out of a total revenue of $41.5 million in 1861) had been the major source of government revenue with some receipts from sale of land and miscellaneous sources.  This would be inadequate.  After steady growth over the four years of war, Federal government expenditures would reach $1,297.6 million with $1,031.3 million going to the Army, $122.6 million to the Navy, and $77.4 million in interest on the national debt.  New taxes were brought in to meet the outflow; a new bureau of Internal Revenue (as opposed to external revenue, i.e., customs duties) was created to manage the collection of a series of new excise taxes and the new income tax.

The principal excise taxes were for alcohol, tobacco, a range of stamp duties, a manufacturers’ tax, and an estate tax.  The stamp duties consisted of obligatory stamps to be affixed to various legal documents such as telegrams, bank checks, foreign exchange agreements, bill of lading, contracts, power of attorney, life insurance, and more plus the familiar playing card tax stamps.  By 1865, these internal revenue collections were nearly three times as large as the customs duties; trade had been depressed by the war.  With time, the economy recovered and by 1875 customs duties again accounted for about half of all revenues, but internal taxes now accounted for about 35% to 40% of all revenues.

At this point, the burden of paying down the national debt shifted.  The estate tax, never very significant, was ended immediately.  The manufacturers’ excise taxes were trimmed to insignificant amounts.  The tobacco, alcohol, and stamp taxes, however, stayed.  The Federal government ran a surplus every year between 1866 and 1893.  In some years, the surplus was equal to a third of the total revenues of the government.  Either the tariff could have been reduced or the excise taxes cut and the government budget still would have balanced, but the debt would not have decreased in absolute terms..  Since most of the excise taxes were on consumable items, they were regressive.  The tariffs were kept high to protect industry.  With the Treasury taking back the Greenbacks, there was a long deflation in the country and the excise taxes hit individuals ever harder.

Growth in the economy would have reduced the actual burden of the 1865 national debt over time.  The decision to pay down the debt in absolute terms, however, was one more burden of the war that the lower orders of the society had to bear more heavily than others.

Graph: the growth of revenues:

Total US Gov’t Revenue 1850-1900

Graph: internal revenues and excises:

Internal Revenues by type 1860-1900

[Objects: Revenue stamps for Probate of Wills and for Playing Cards.]    

The price of war

“Or what king, going to make war against another king, sitteth not down first, and consulteth whether he be able with ten thousand to meet him that cometh against him with twenty thousand?”  Luke 14:31

Modern war may be seen as a contest between two economies to see which can sustain the larger shock and loss of its capital stock.   If that proposition has any validity, then the cause of the Confederacy was particularly ill-planned for fighting a war.  The basic challenge was two-fold.  The states that seceded had great personal wealth, but it was of a very illiquid type being in the form of land and associated buildings, and (odiously) the price put on human chattels.  This was not financial wealth (and there was very little money) that could be converted into materiel for armies or used to pay taxes or purchase bonds.  Further, the war strategy of the government was to make the burden on the population caused by taxation to be as light as possible.  Instead, the government would rely on borrowing.  [Speech by Alexander H Stephens, July 11, 1861.] 

Indeed, there was a war tax approved in 1861 and no other for the next two years.  When, in the later years, some attempt was made at taxation, it often became a payment in kind.  Some taxes led to confiscation.  (Curiously, there was no tax on slaves as that would, by the new Constitution, require a census which wasn’t high on the priority of the government.  This meant, in practical terms, that large slave holdings could pay a lower tax rate than merchants.)  Some bond issues became obligatory for taxpayers to purchase.  Nor could the Confederate treasury benefit from the tariff its Congress approved because trade suffered from the US blockade.  Over all, it is estimated that the Confederacy raised about 5% of its revenue over the four years from taxes and about 35% from loans.  The remaining 60% was paid through the printing press; there really was no alternative.  As a result, inflation of the few goods available was rampant.  The amount of gold that could be commanded by $1.10 in Confederate notes on the Richmond exchange in May 1861 would require $60 to $70 of the same notes in April 1865.  That is an overall inflation of about 7000%.

In contrast, the United States paid for much of its war costs by taxes – excises, tariffs, and a new income tax – and by borrowing.  There already was so much financial wealth, particularly in the Northeast, that these two measures could carry much of the burden.  The borrowing was principally carried out by Jay Cooke and Co. who managed what might be called syndication: they took responsibility for selling the debt to investors rather than leaving the task directly to the Treasury.  The same estimates of means of finance for the US put the share raised by taxes at 21% of its revenues and that from borrowing at 65%, leaving only 14% to be financed by money such as the new Greenbacks.  The gold commanded by $1.01 in New York in April 1861 could be bought with $1.46 in April 1865.  This is a much lower rate of inflation.

Graph: Gold prices in US and CSA notes, data from Todd, Confederate Finance:

Relative gold prices CSA&USA 1861 to 1865

Taxation may not be politically popular – LBJ tried hard to avoid it until Wilbur Mills insisted on a tax surcharge to pay for Vietnam – but, it does restrict the damage to an economy that wars can bring even without invasion.

[Object: Confederate $100 note of February 1864.]  

Easy to do, if a bit costly …

NewScan004Here is a form letter from the Treasury Department sent to Mr. Albert Blanchard of Richmond, Indiana, an investor in government bonds in 1861, informing him that his deposit has been received.  Two things stand out about this document.  One is that it is October 1861 with the crisis of the American Civil War now going on for six months and the Treasury still needs time to prepare securities to fund the effort.  Second is that it should be noted that these securities are drawing a coupon of 7 and 3/10s percent.

I do not yet understand why the printing presses and the clerks are delaying the issuance of the debt securities.  Perhaps because the money is already in the Treasury there was no need for rush.  The letter states that two certificates of deposit have been issued in Mr. Blanchard’s name.

Further, the 7 3/10s rate is odd; one usually expects something in eighths or sixteenths as customary for financial markets and thirty-secondths have held a place in bond markets for a long time.  Decimalization is fairly new to finance; the SEC ordered US securities markets to begin pricing in decimals only on April 9, 2001.  Using tenths in a bond transaction was uncommon.  As to the 7.3% rate, the first impression may be that the market already had doubts about the success of the Union cause.  It had not been an auspicious summer in the East; the United States’ armies had been embarrassed, if not defeated on the field.  Surely, I thought, the market was expressing concern.  The attached table of public debt for 1869 lists some outstanding issues from before the war.  These include a 15-year bond issued in 1858 at 5%, a 10-year bond issued in 1860 at 5%, and two 20-year bonds issued early in 1861 at 6%.  The jump in bond rates might just show the market’s nervousness, particularly as the 7.30 (as it was known) was only for three years.

I was wrong.  I forgot that these were not rates set by market forces at an auction.  (Setting the coupon rate during an auction is a quick and efficient way to grasp how the market feels about a country’s creditworthiness relative to other countries or assets.  That is what the US Treasury does today.)  In 1861, however, the coupon was chosen and then the subscription efforts started.  Choosing the coupon first requires a good feel for the market.

The rate in this case was chosen by the Secretary of the Treasury and it was done for calculation convenience: the 7.3% rate comes to 2 cents per hundred dollars per day.  The issue was actually quite popular and known as seven-thirties.  (The previous issue, the five-twenties, were a twenty year bond that paid at six percent in gold.)  Other issues during the war were generally at 6%.

It is the popularity that really troubles me.  If an issue is too popular, it seems likely that the coupon rate was set too high.  After all, interest is a zero sum game: every interest payment made to the lender is a cost to the borrower. The happier the lender, the worse off the borrower.   Some countries try to limit their borrowing costs by intervening in some way in their auctions – refusing offers that are too low (i.e., at a high interest rate) or by choosing the coupon first.  These practices may help the debtor control their borrowing costs, but that certainty is purchased at the risk of not being able to sell the whole issue and, thus, making the amount raised uncertain.

For the price, maybe the Treasury should have hired a few more clerks who were good in arithmetic.


Table: Statement of Outstanding Debt, March 1869



[Object: U.S. Treasury letter of Oct 14, 1861 acknowledging deposit of funds to purchase Treasury Notes at 7.3% as soon as securities are prepared.]

Sometimes, the world intervenes …


This 20-year Greek bond of 1922 paid interest annually in April.  Unfortunately, Greece was invaded by the Nazis in March of 1941.  I suspect that may be why the 1941 and 1942 coupons are still there.

If one collects old securities or bonds (a hobby called scripophily), you see this fairly often.  I have a Russian Imperial bond  for 125 gold rubles (equivalent to 500 francs, or 404 Imperial German marks, or 239 Dutch guilders, or 360 Danish krona, or $96.25 in gold, or 19 pounds sterling 15 shillings and 6 pence) from 1894 with the coupons intact from May 1918 forward.  I guess they clipped the Nov 1917 coupon but found no bank to cash it.  The story is the same with a bond issued by St Petersburg in 1908.  Finally, a friend gave me a lovely bond issued by Montevideo after 1943.  It still carried most of its coupons.  As the friend explained, inflation became so severe that the amount of cash the coupon paid out was less than the transit ride to the bank to cash it.

In a world of perfect information, no inflation, and no uncertainty, the yield on a bond would be just enough to induce the holder to put off enjoying, or consuming, whatever could be bought today with the funds invested until that future date when, with interest, the funds available will be greater and, expectedly, consumption can be greater.  Pure interest measures the trade-off,  between pleasure today and pleasure in the future.

Of course, it is an imperfect world.  Inflation consumes savings by eroding the purchasing power of funds that have been reserved for the future;  to put off consumption that could be enjoyed today for a larger amount in the future requires that the earnings on the deferred consumption maintain their purchasing power despite inflation.

That is why the yield, i,  required of a bond is not just the sum or the two components real interest, r, plus expected inflation, p, but the product of them

i = (1 + r)(1 + p) – 1 =  r + p + rp.

In a similar manner, there is also a risk, d, that some event will intervene that keeps the borrower from paying maturing principle or interest at all.  The premium applied by the lender to cover this risk  must also apply to all the elements of the payout.  Therefore, the yield required should be represented as

i = (1 + r)(1 + p)(1 + d) – 1 =  r + p + rp + d + dr + dp + drp.

This is usually just shortened to i = r + p + d, but that assumes those second-order terms are not significant.  The burden of calculation is slight.  I would include them.  After all, the world sometimes intervenes.

[Object: Kingdom of Greece 6.5% 20-year bond of 1922.  One coupon per year payable in April.]