Any Bonds Today?

Can you imagine Julia Roberts and Gwyneth Paltrow helping the US government sell bonds or Jay Z and Justin Timberlake composing songs about Treasury bills? It would not be the first time Hollywood stars or famous musicians tried to help the government sell its debt.

The last time the US government had an enormous load of debt, it used Hollywood stars to help sell government debt. The Treasury Department conducted a massive public relations campaign through radio, newspapers, and film. During World War II, war bond rallies were held throughout the country, and Hollywood stars such as Bette Davis and Rita Hayworth traveled around the country to promote war bonds. The great Irving Berlin even wrote a song titled “Any Bonds Today?” and Berlin’s tune became the theme song of the Treasury Department’s National Defense Savings Program.

The government also enlisted cartoon characters, actors, comedians, and musicians to encourage people to pay income taxes. Donald Duck told viewers it was their “duty and privilege” to pay income tax. Abbott and Costello appeared in advertisements to get people to pay taxes, and Irving Berlin wrote songs not only about bonds but songs about taxes like “I Paid My Income Tax Today.”

While the war bond and income tax drives garnered all the press, the real reason the US was able to borrow so much and with so little burden had nothing to do with the glitz and glamor of movie stars. The US government borrowed easily because the Federal Reserve printed money to keep interest rates low. Borrowing is very easy when a central bank has your back.

How did it work in practice? As is the case today, the Treasury wanted to borrow cheaply then, and the central bank was happy to accommodate. In 1942, after the United States entered World War II, the Federal Reserve officially agreed to fix interest rates on government bonds at a low level. To maintain the pegged rate, the Fed was forced to give up control of the size of its balance sheet. Unsurprisingly, the Fed bought and held all available short-term US treasuries and almost all long-term government bonds.

The costs of paying for World War II pushed the national debt up sharply, from around 40% of GDP before the war to a peak of nearly 110% as the war ended. But a combination of strong economic growth, tight fiscal policies, and financial repression brought the debt back below 50% of GDP by the late 1950s. (Currently our government debt has reached about 90% of GDP and continues climbing very sharply.)

During the war years, the Federal Reserve pegged long-term interest rates at extremely low levels so the government wouldn’t have to pay much to fund itself. To make sure that inflation didn’t spike, the government instituted wage and price controls. After the war, the price controls disappeared and inflation rose very quickly, averaging about 6.5 percent annually from 1946-51. By the postwar price peak nine years later, wholesale prices had more than doubled, and the stock of money had nearly tripled.

Normally, such high inflation would have made it much more expensive for the government to borrow money. But after being pressured by the Treasury, the Federal Reserve agreed to keep on pegging long-term government bond yields at 2.5% until the spring of 1951, when the Federal Reserve finally refused to print money to keep bond yields low. Because of the coordination between Federal Reserve and the US Treasury, real yields on government bonds were very negative during the years following World War II. With negative real yields, borrowers win and lenders lose. The clear winner was the US government, and the loser was anyone who bought and held US bonds. The combination of very low government bond borrowing costs and high inflation ate away a sizable chunk of the government’s debt burden.

The same thing is happening today in almost all government bond markets around the world. Governments are winning, and investors are losing. The Federal Reserve is helping the Treasury to borrow cheaply while the government expands its deficit spending and debt accumulation. Using inflation and low bond yields this way to reduce government debt is called financial repression.

The government and central banks also contribute to higher inflation by pretending inflation is always under control. For example, throughout the Greenspan and Bernanke years, the Fed consistently chose to focus on lower inflation measures whenever doing so suited the central bank. You can see this in the semiannual monetary policy reports to Congress, specifically in the inflation forecasts made by the members of the Federal Open Market Committee. Until July 1988, inflation forecasts used the implicit deflator of the gross national product, but then the Fed switched to the Consumer Price Index. In February 2000, the Fed replaced CPI with the personal consumption expenditures (PCE) deflator. Thus from July 2004 onward, inflation forecasts have employed the core PCE deflator that excludes food and energy prices. Using lower and lower, less comprehensive estimates for inflation has allowed the Fed to pretend that it is meeting its mandate – but by ignoring high inflation readings. In the meantime, interest rates have been kept too low, and the inflation rate has consistently remained above the Federal Funds rate.

But measuring inflation is not so easy. The vast majority of readers have no idea about the rather contentious nature of the debates that go on in academic conferences about arcane topics such as the minutiae of how to measure some minor aspect of inflation. Passions run deep. Careers are made. Once you delve into how things are actually done, you realize that what we think of as an inflation number is actually an approximation of an idea the very definition of which can change over time.

Your perception of inflation (and everyone else’s) has a very close relationship to how stock markets perform over time. Indeed, one of the questions we are both regularly asked wherever we speak is something along the lines of “What do you think inflation or deflation will be?” And the answer is not easy: it depends on a number of factors that vary from country to country.

In general, the trend for the last 75 years has been one of inflation. Sometimes, in some countries, inflation has spun out of control. At other times you see outright deflation. Neither one promises good times for investors. Ever-falling inflation or low inflation is the best environment for investing. But given the paramount importance of the inflation/deflation debate, we need to briefly investigate what inflation is and is not.

There has been a great deal written about the difficulty of measuring inflation and about the potential manipulation of inflation statistics over the last 30 years. John Williams of ShadowStats is the most-noted proponent of the position that inflation is running well above the current US government’s number of 2% (for the 12 months ending February 2013).

Employing the methodology that was used in 1980 under the Carter administration, inflation would currently be about 9.6% (see chart below). Using the government methodology from 1990, inflation today turns out to be a little under 6%.

The topic of those alternate inflation numbers comes up often at our tables of conversation. Generally it seems to be clear that the methodologies used in 1980 and 1990 are visibly, patently, demonstrably wrong. If inflation were now at 9.6%, then interest rates should be closer to 12% and not the 1.75% we see on the 10-year Treasury today (more on that topic in a minute), no matter what the Fed wanted, unless they were willing to monetize not only new debt but any existing debt that got rolled over as well. Over time, markets respond to actual inflation and not government statistics. Argentina’s government can state that inflation is “only” 10%, but the market thinks it is 30% and rising.

The government calculation of inflation in 1980 or 1990 was the best they could do at the time. Gentle reader, it was a government calculation. There is nothing ex cathedra about either methodology. In religious terms, neither rises to the stature of the original Greek documents or the Latin Vulgate Bible. Changing the words (the equations) in economics should not be seen as somehow equivalent to changing the fundamental documents of a religion. There is nothing sacred about 1980 CPI methodology, and in fact we can look at it empirically and understand that it was pretty flawed.

You might have some personal investment bias (read “quasi-theological reason”) to want inflation to be high. But that is a belief system. It is one form of faith-based economics (It is not a large stretch to suggest that most economic schools require of their adherents a measure of faith and belief). Expectations of high inflation are for some people a basic tenet of their belief system. Saying there is only a little inflation must therefore be a government manipulation.

We must constantly be comparing our assumptions against what we observe in the real world, in order to discern where our models, with their built-in assumptions, bias our conclusions about what the data says.

If you think overall general inflation is high, then you have to think the entire world is delusional. (Note: your personal inflation rate may be much higher than 2%.) G-7 interest rates are at an all-time low today. That can and will change; but right now the bond market does not see inflation as a problem anywhere in the developed world, although Japan has now made what must be their 10^th vow in the last 20 years to create inflation. This time, they may actually (for them, catastrophically!) succeed. For now, however, deflation and deleveraging are the order of the day.

If we had kept the methodology used until 1980 for calculating the Consumer Price Index and then used that number to adjust Social Security and government pensions, the US government would be bankrupt today. Social Security would have gone negative in the 1990s and tripled in cost in the last 12 years (compounding at 10% can do that). Now, those of you living on Social Security might think a tripling of payments is appropriate, given what has happened to your budgets, but younger taxpayers would hasten to differ. (Note: we are not arguing that SS provides a livable income at current levels. Different topic for another paper.)

All this is not to say that today’s inflation methodology is correct or gives us a number that is accurate. It is simply better than the methodology used in 1980 – but it is still just a statistical method that tries to reach for the impossible, all-illumnating star of reliability and finally has to settle for accuracy in general at the risk of imprecision in the particulars. We will be able to look back in 15 years to see how well we are doing today at measuring inflation. The real surprise would come if we don’t change methodologies at least a few more times between now and 2030.

It is hard to argue with people who point out that prices and the cost of living are going up faster than government-reported inflation reflects. We can all see prices rising. Food, energy, tuition (try managing all that for 30 years with seven kids!) – they’re all going up. If we had used actual home prices in the CPI, inflation would have been seen as very high in the middle of the last decade. Instead, we seemed to be flirting with deflation; and if we used housing prices in 2008-2011, we would certainly have had government-reported deflation. In place of home prices, the Bureau of Labor Statistics decided to use something called Owners’ Equivalent Rent a few decades ago; and it is the largest part, a full 24%, of the CPI. Something called hedonics is probably the most contentious part of the CPI calculation. The BLS says, “The hedonic quality adjustment method removes any price differential attributed to a change in quality by adding or subtracting the estimated value of that change from the price of the old item.” This is not as mysterious as it sounds. When, for example, you replace your old computer with a new one, paying roughly what you did before, the new model you buy is always faster and more powerful than the old one. The BLS says you are getting more for your dollar; therefore the price fell even if you paid as much or more for the new computer. Opponents say hedonics can be used to hide “true” inflation.

We do know that a lot of items have in fact gone down in price and up in quality or capacity. Cell phones are a good example. And the cost of using cells may be ready to really fall. There is a full smart phone that uses a major carrier and Wi-Fi in combination now on the market for $20 a month for all the voice, data, and text you can eat. It works on Wi-Fi in Asia, in Europe, and in the middle of the Andes. You pay basically nothing for 10 or 15 or 40 hours a week of talk time, and people can call you anywhere in the world using a local US number if you are connected to Wi-Fi.

Most egregiously for many, the CPI also does not take into consideration income taxes. For a number of people, taxes are their largest source of inflation!

Yet all of us here in the US are governed by the same people in Washington, and they define inflation in their own way, via the Consumer Price Index and various related benchmarks. Because CPI tries to find a national “average” inflation rate, it is almost by definition inaccurate for any given person, family, business, city, or state. CPI is the least common denominator, a “one size fits all” coat that in reality fits no one very well. (For the record, all the data used to calculate inflation is public. You can calculate inflation for your own local area if you have nothing better to do. In fact, the entire methodology is public, if a little dense.)

Given the acknowledged limitations of the CPI, we nevertheless use it in myriad ways. It governs cost-of-living adjustments for Social Security beneficiaries, government employees, and many labor union members. CPI is baked into the general cake, even though we know it is an imperfect fit in almost every situation.

As a result, some people get raises when their cost of living drops, while for others the cost of living rises faster than their income does. Is this fair? No. Is there a better way? We don’t know what it would be. There are hundreds of smart people who build entire careers trying to answer that question.

Other inflation measures exist, but they all have their own limitations. Three Federal Reserve Bank regions calculate their own versions of CPI. The Federal Reserve itself prefers to look at something called PCE, or Personal Consumption Expenditures, a measure which uses chained dollars rather than a fixed basket as the CPI does. Since 2000, the Federal Reserve has used PCE in its reports to Congress about expectations for inflation.

In explaining its preference for the PCE, the Fed stated:

The chain-type price PCE index draws extensively on data from the consumer price index but, while not entirely free of measurement problems, has several advantages relative to the CPI. The PCE chain-type index is constructed from a formula that reflects the changing composition of spending and thereby avoids some of the upward bias associated with the fixed-weight nature of the CPI. In addition, the weights are based on a more comprehensive measure of expenditures. Finally, historical data used in the PCE price index can be revised to account for newly available information and for improvements in measurement techniques, including those that affect source data from the CPI; the result is a more consistent series over time. (“Monetary Policy Report to the Congress,” Federal Reserve Board of Governors, Feb. 17, 2000)

Contentious? You bet! PCE and other chained inflation numbers generally yield lower inflation figures, which is why many in Congress (and the AARP) think the “chained dollars” amount to some sort of conspiracy to defraud seniors on Social Security. CPI is used to calculate adjustments for income taxes. If it is too low, then incomes rise faster in real terms than cost adjustments do, and that acts as a tax increase even as your pension is adjusted lower. But if inflation is calculated too high, then taxes are lower than they would otherwise be and the costs of Social Security and pensions are higher. Talk about using a sledgehammer to fine-tune a highly developed economy. Even small miscalculations will add up over time to large losses for someone. Ouch!