Dr. Janet Yellen

[Revised October 22 to correct the misimpression that the Fed is wholly responsible for the increase in inequality.  Thanks to Mark Dow @Mark_Dow for pointing out this issue.]

On October 17, Fed Chair Janet Yellen gave the keynote address at the Federal Reserve Bank of Boston’s 58th Conference on Inequality of Economic Opportunity.  The keynote was at 8:30 am, perhaps indicating that Dr. Yellen would rather not have had many attendees.  Indeed, given the content of her talk, such fears appear justified.  Dr. Yellen decries wealth inequality, fails to mention Fed’s role.  I hasten to add that the Fed’s policies over the past six years are just one factor causing inequality to increase.  Technological change, specifically the improvements in the speed of information transmission, are at least as important and probably a bigger factor.  That’s certainly the case in the long run.

Dr. Yellen’s idea is analogous to an old joke →

Exactly.  Dr. Yellen thinks there is too much income and wealth inequality in the U.S. today.  First, I have to note that this falls into “normative economics,” statements made by economists based at least partly on their opinions.  Now my opinions about, say, the minimum wage are likely to be more informed than years.  I get paid to keep up with the research.  But when it comes to opinion, yours is as good as mine — and both are as good as Dr. Yellen’s.

Second, what in the world is the Fed supposed to do about inequality?  At the moment, they can’t even perform one of their two main tasks: stimulating employment.  Oh, well, at least inflation is under control.

Until the Fed has solved the problem of the “zero lower bound” on interest rates, perhaps they should concentrate on their core mission and leave inequality debates to others.

Here are some excerpts from the Wall Street Journal coverage:

BOSTON—Federal Reserve Chairwoman Janet Yellen said rising inequality of wealth and income in the U.S. was impeding the economic mobility at the heart of American values.

“The extent and continuing increase in inequality in the United States greatly concern me,” Ms. Yellen said to a conference on economic opportunity and inequality sponsored by the Federal Reserve Bank of Boston. “I think it is appropriate to ask whether this trend is compatible with values rooted in our nation’s history, among them the high value Americans have traditionally placed on equality of opportunity.”

…

Ms. Yellen didn’t address Fed critics’ arguments that the central banks’ bond buys and zero-interest-rate policies have contributed to inequality by bolstering prices of assets such as stocks, which are primarily held by wealthier Americans.

The Fed counters its policies are aimed at boosting the overall economy, and thus helps lower-income Americans who are more likely to have high levels of debt.

It happens that the Fed critics are exactly right. While interest rates on debt held by those in the bottom 80% may be slightly cheaper, much of that debt is probably credit cards. Those interest rates are still way, way above 1%. Mind you, I have no problems with those rates. These are unsecured loans to borrowers, with no collateral at all. The high interest rates reflect the risk of those loans. Thus banks are able to borrow at low rates and lend to the poor and middle class at fairly high rates.  Again, I must add that these low interest rates are a contributing factor, not the sole reason for the growth of inequality.

Theory and a Hypothesis

Decades ago I studied monetary theory with Prof. Karen Johnson.[1] I learned this lesson over and over:

All assets are substitutes for each other. But some are better substitutes than others.

Central banks around the world have kept interest rates very low. The Fed, via their “quantitative easing” program has kept short-term U.S. rates near zero and longer-term rates in the 2% – 3% range. But portfolio managers need to turn a profit. They seek yield. If they can’t find it in U.S. government securities, they look elsewhere. Junk bonds, high-yield foreign bonds (and don’t forget Puerto Rico), and, yes, stocks. Probably real estate, too. Junk bond yields are in the 5% to 9% range. High-yield foreign bonds are around 4.5% – 6%. Where, oh where can they go?

Stocks, naturally. The stock market has boomed. There is widespread agreement among those with a clue that even a hint at higher interest rates would cause a selloff in the U.S. stock market and probably some other countries, too.

While U.S. stocks are more broadly owned than is generally believed (mutual funds, retirement accounts, etc.), it remains true that wealthier households own more shares of more stocks than less wealthy folks. And their wealth has risen accordingly.

Hence my hypothesis: the Fed’s policies have caused, at least in part, the increase in wealth inequality. I will not discuss the many, many other government programs that have the same effect. But I will note that the residents of the zip code 94027 are the largest purchasers of Tesla automobiles (price: $60,000 up). That zip code is for Atherton, CA, one of the wealthiest suburbs of San Francisco. I’m certain Atherton’s proud Tesla owners are also enjoying the tax benefits of their purchases.

Testing the Hypothesis

Using data from the Census I developed six measures of inequality.[2] They were:

1. Percentage of U.S. households in the top 20 percent of income.
2. Percentage of U.S. households in the top 5 percent of income.
3. Median net worth (in dollars) of the wealthiest 20 percent of households.
4. Percentage of households with a net worth of $500,000 or more.
5. The ratio of mean to median net worth of the wealthiest 20 percent of households.
6. The Gini coefficient for the U.S.

For the independent variable I used the yield on ten-year U.S. government notes. All data is annual. The independent variable and the Gini data are from the FRED database at the Federal Reserve Bank of St. Louis. All other data is from the Bureau of Census. For the ten-year note, I have data from 1980 through 2013. Other series vary both in length and continuity as noted in the following table.

All coefficients of the interest rate are negative, indicating that lower interest rates increase inequality. Four of the five are statistically significant. (The lone exception is my constructed variable, the ratio of mean to median net worth.) Goodness-of-fit varies widely, but is generally better for the series with higher n’s.

Conclusion

Decades ago, Prof. Mike Hurd drummed another valuable lesson into my brain. Without some extensive and very difficult work, you cannot calculate the power of a statistical test. Unfortunately, that’s what you need to do to “confirm” a null hypothesis. As things stand, all I can say is that I failed to reject my null hypothesis.

For those interested, I have put my Excel workbook and the SPSS files into a zip file. Click here to download it.

[1] Dr. Johnson has now retired after a successful career at the Federal Reserve. Naturally she is not responsible for any of my errors — here or elsewhere!

[2] In some cases, my “development” consisted simply of copying the Census data into a format consistent with my original dataset.