Economics in central banking: Enrique Martínez García, the Federal Reserve Bank of Dallas

When Enrique Martínez García and his team first started developing metrics to track explosive growth in house prices, there was only one major example of a boom period that showed up in the data: the run-up to the global financial crisis (GFC). The fact that there are now two, underlines how his breakthrough research is delivering critical policy insights. It also implies the global financial system could be in serious danger.

Martínez García’s work has its roots in the GFC. He joined the Federal Reserve Bank of Dallas straight out of grad school under Mark Wynne, now associate director of research at the regional bank, responsible for the 11^th Federal Reserve district. Wynne was recruiting staff to the Dallas Fed’s Globalization Institute, formed to explore the implications of globalisation on monetary policy. With the crisis as a backdrop, “a natural research question was to develop a better understanding of what was going on in housing markets, both in the US and around the world,” says Wynne.

The institute was still some way from being able to develop indicators of housing booms. Two key pieces of research provided the necessary foundation. First, Martínez García and colleagues Adrienne Mack and Valerie Grossman established the International House Price Database (IHPD), which now tracks 25 major jurisdictions on a quarterly basis. A December 2011 paper sets out their methodology for gathering the data, which stretches back to 1975. Martínez García emphasises the additional contributions of Jarod Coulter, Kostas Vasilopoulos and Lauren Spits to the IHPD.

Second, Peter Phillips, an econometrician at Yale University, developed the generalised supremum augmented Dickey-Fuller (GSADF) technique for identifying multiple bubbles in a data series. Along with Shu-Ping Shi and Jun Yu, Phillips applied this to the S&P 500 to show it is possible to put a time stamp on the start and end of bubbles, even where there are many in a series that keep blowing up and collapsing. “The difficulty arises from the complex nonlinear structure involved in the multiple breaks that produce the bubble phenomena,” the authors write. They note that solving this problem is especially important if the method is to be used for policy – you don’t want to be recommending a costly intervention where it might ultimately prove unnecessary.

Martínez García saw the potential in uniting these two research streams. By this point, the Dallas Fed team was collaborating with economists at the UK’s University of Lancaster, led by Efthymios Pavlidis and Iván Payá, to examine the data. This yielded the 2016 paper, Episodes of exuberance in housing markets: in search of the smoking gun. The paper not only applies the GSADF test to the Dallas Fed’s housing market data but also extends the technique to make it applicable to panel data, allowing the researchers to explore the insights contained in their large cross-section of housing markets around the globe.

A further insight was to use a broader set of indicators to cross-check signals coming from real house prices. The IHPD team looks at signals of exuberance based on the price-to-income ratio as a measure of housing affordability, and at the price-to-rent ratio, which is broadly analogous to the price-to-earnings ratio for stocks. Alongside a run-up in real house prices, signs of exuberance tend to show up as rapid declines in affordability and increases in the price-to-rent ratio. If all three are flashing red, the problem is likely to be acute.

As daunting as the GSADF test may sound to a non-specialist, the results are very easy to interpret. Martínez García compares it to a thermometer, which a doctor can use to take a patient’s temperature. “It is very simple to illustrate,” he says. “You have a critical value, just like you have a reading of what normal temperatures are, and you have a reading.” Above the critical value, markets are shifting into “explosive” mode, where prices are likely to be unmoored from fundamentals and can rise rapidly and unsustainably. The metric highlights the build-up of imbalances ahead of the GFC, showing “widespread and synchronised episodes of exuberance” across housing markets worldwide.

A question left unanswered by the 2016 paper was whether the method would prove useful in detecting the next episode of exuberance. That changed as the Covid-19 pandemic struck, triggering not the collapse in asset prices that many feared, but rather a surge. Many housing markets across the globe crossed their critical thresholds, including the US. By the first quarter of 2022, the US was in the danger zone across all three of the key ratios. Martínez García sounded the alarm in an article for the Dallas Fed in March, and again in November.

A few days after the November article came out, Fed chair Jerome Powell issued an unusually blunt warning in remarks at the Brookings Institution. “Coming out of the pandemic, rates were very low, people wanted to buy houses, they wanted to get out of the cities and buy houses in the suburbs because of Covid,” he said. “So you really had a housing bubble, you had housing prices going up at very unsustainable levels and overheating.”

Powell did not cite his sources. But Martínez García’s work has been cited widely in academia, as well as informing private-sector analysis and being picked up widely in the press. It has underpinned reports published by central banks including the European Central Bank, Bank of England and Sveriges Riksbank, as well as the International Monetary Fund’s World Economic Outlook. His March 2022 article for the Dallas Fed is the most read in the website’s history.

Economists who have followed Martínez García’s work praise the “state-of-the-art” econometrics used to generate the results, as well as the fact that the team makes its data and code free for public use.

“I think very highly of Enrique’s work,” says Prakash Loungani, assistant director at the International Monetary Fund. “He has worked hard to assemble the data and use it to provide indicators of overvaluation in a rigorous and transparent way.”

The method has its limits. Just as a patient can run a fever for many reasons, housing markets can be hot due to a range of factors, some more dangerous than others. It is also hard to say exactly when or if a bubble will burst. But it allows policy-makers to investigate further and intervene if necessary. Martínez García sees it as critical to communicate early when a bubble is emerging. He notes bubbles are driven by expectations of further price gains, so if you can puncture them before they get out of hand, the air can come out of the market without a major crash.

This now appears to be happening, at least in the US. Higher interest rates and a surge in house building (which is now tailing off) led prices to peak in Q2 2022. Monthly data is available for the US, allowing policy-makers to track developments faster, and Martínez García’s team is looking into even higher-frequency indicators. He notes the exuberance indicator is helpful on the downside too, helping policy-makers to gauge how quickly markets are cooling, which is an important input for monetary policy.

Wynne sums up the impact: “In short, I think the IHPD and the related exuberance work have been very important to the Dallas Fed, the broader Federal Reserve System and central banks around the world, and we plan to continue to support this work going forward.”

The Central Banking Awards 2023 were written by Christopher Jeffery, Daniel Hinge, Dan Hardie, Joasia Popowicz, Ben Margulies, Riley Steward, Jimmy Choi and Blake Evans-Pritchard.