Estimating and Evaluating a Fed Policy Rule with ChatGPT

Estimating and Evaluating a Fed Policy Rule with ChatGPT*


In a collaborative effort with the AI model ChatGPT, I’ve embarked on an analytical exploration of the Federal Reserve’s monetary policy decisions over the years. Leveraging a model inspired by Greg Mankiw’s approach, we’ve incorporated key economic indicators, including the 10-year government bond yields, to provide a more nuanced understanding of the Fed’s policy rule.

Integrating Bond Yields into Our Model

The inclusion of 10-year government bond yields in our OLS regression model, alongside unemployment and core CPI inflation, offers deeper insights into how the Fed’s policy decisions have been influenced by long-term market expectations and economic conditions.

Critical Periods in Federal Reserve Policy

Our analysis focuses on several pivotal periods in the Fed’s history, evaluating its responses to major economic challenges:

  1. Early 1980s – Inflation Battle: Under Chairman Paul Volcker, the Fed’s aggressive interest rate hikes were crucial in taming the high inflation of the 1970s. Our model suggests these measures were necessary, though they came with significant economic trade-offs.
  2. Tech Bubble Burst Around 2000: In response to the dot-com bubble’s burst, the Fed’s rate adjustments aimed to stabilize the economy. Our analysis indicates these were generally effective, but they possibly laid the groundwork for future market imbalances.
  3. 2008 Financial Crisis: The Fed’s response to the financial crisis, including dramatic rate cuts and quantitative easing, was a bold departure from previous policy norms. Our model reflects these actions as crucial in mitigating the crisis’s impact, though their long-term implications remain a subject of debate.
  4. Pandemic Response: The COVID-19 pandemic saw the Fed implementing near-zero interest rates and extensive monetary support. According to our model, this response was aligned with the unprecedented nature of the crisis, but a full evaluation will require more time and data.

Graph 1: Federal Funds Rate and Major Economic Events (1980-2020)

Current Monetary Stance

With the recent economic disruptions, the Fed faces a challenging balancing act. Our model, which now includes bond yields, suggests a cautious approach as the economy recovers from the pandemic’s impacts.

Graph 2: Actual vs. Predicted Federal Funds Rate (Including Yields)

Concluding Reflections

This comprehensive analysis, enriched by the inclusion of bond yields, underscores the intricate dynamics of the Federal Reserve’s policy decisions. Understanding past trends and current factors provides valuable insights for anticipating future policy directions.

….

* The entire content of this blog post, including the modeling, was crafted by ChatGPT under my direction and guidance. In this process, I utilized ChatGPT as both a research assistant and a ghostwriter. ChatGPT has significantly aided in generating new ideas, enhancing my productivity. However, it’s important to note that ChatGPT’s standalone output often lacks significant interest. The true value emerges from the interactive synergy between ChatGPT and the user – that’s where the magic happens. For the data analysis presented here, I uploaded information from the St. Louis Fed FRED database into ChatGPT. (A plugin for this would be highly beneficial!)…even this was written by ChatGPT.

Where is inflation going?

A couple of weeks ago I was invited to do a talk on the topic “Where is inflation going?” at a webinar organised by the Institute of International Monetary Research (IIMR) at University of Buckingham.

Hosted by IIMR Director Damian Pudner in collaboration with the Vinson Centre for the Public Understanding of Economics and Entrepreneurship at the University of Buckingham.

As always it was a great pleasure for me cooperating with the IIMR and presenting my views on monetary matters.

My view on where inflation is coming from is summarised in the picture below.

And my answer to where inflation is heading is summarised in the graph below.

But there is more to it than that so have a look at the webinar for yourself.

Tight money = low yields – also during the Great Recession

Anybody who ever read anything Milton Friedman said about monetary policy should know that low interest rates and bond yields mean that monetary policy is tight rather than easy. And when bond yields drop it is normally a sign that monetary policy is becoming tighter rather than easier.

Here is Friedman on what he called the interest rate fallacy in 1997:

“After the U.S. experience during the Great Depression, and after inflation and rising interest rates in the 1970s and disinflation and falling interest rates in the 1980s, I thought the fallacy of identifying tight money with high interest rates and easy money with low interest rates was dead. Apparently, old fallacies never die.”

Unfortunately the old fallacy is still not dead and it is still very common to associate low interest rates and low bond yields with easy monetary policy. Just think of the ECB’s insistence that it’s monetary policy stance is “easy”.

In my previous post I demonstrated that all the major changes in the S&P500 over the past four years can be explained by changes in monetary policy stance from either the ECB or the Federal Reserve (and to some extent also PBoC). Hence, it is not animal spirits, but rather monetary policy failure that can explain the volatility in the markets over the past four years.

What holds true for the stock market holds equally true for the bond market and the development in the US fixed income markets over the past four years completely confirms Milton Friedman’s view that tighter monetary policy is associated with lower bond yields. See the graph below. Green circles are monetary easing. Red circles are monetary tightening. (See more on each “event” in my previous blog post)

I think the graph very clearly shows that Friedman was right. Every time either the ECB or the Federal Reserve have moved to tighten monetary policy long-term US bonds yields have dropped and when the same central banks have moved to ease monetary policy yields have increased.

Judging from the level of US bond yields – and German bond yields for that matter – monetary policy in the US (and the euro zone) can hardly be said to be  easy. In fact it is very clear that monetary policy remains excessively tight in both the US and the euro zone. Unfortunately neither the Fed nor the ECB seem to acknowledge as they still seem to be of the impression that as long interest rates are low monetary policy is easy. I wonder what Friedman would have said? Well, I fact I am pretty convinced that he would have been very clear and would have been arguing with the Market Monetarists that monetary disorder is to blame for this crisis and we only will move out of the crisis once the Fed and the ECB move to fundamentally ease monetary conditions and adopt a rule based monetary policy rather than the present zig-zagging.

PS See also my early post on the connection between monetary policy and the bond market: Understanding financial markets with MV=PY – a look at the bond market
Update: Scott Sumner just made me aware of one of his post addressing the same topic. See here.

Update 2: Jason Rave has kindly reminded me of this Milton Friedman article, which also deals with the interest rate fallacy.