Researchers across the 12 district banks are racing to find better ways to measure growth, employment and inflation (photo: ThinkStock).

(Bloomberg) – When Janet Yellen decides she’s ready to raise interest rates again, her impetus may be her go-to source on wage trends: the Federal Reserve Bank of Atlanta.

The Fed chair in June cited the district bank’s data tracking wages of individuals from year to year, which showed a robust 3.6 percent increase in the latest period. The Atlanta Fed is also the source of “everyone’s favorite tracking estimate” for gross domestic product, says Neil Dutta, head of U.S. economics at Renaissance Macro Research LLC in New York.

As the central bank tries to understand the U.S. economy, researchers across the 12 district banks are racing to find better ways to measure growth, employment and inflation. On Peachtree Street in midtown Atlanta, 23 economists have developed some of the Fed’s more influential gauges that are shaping the debate about when next to change interest rates.

Related: Fed’s mission is accomplished. So why no change?

“The Atlanta Fed is doing some fascinating research,” said Drew Matus, deputy U.S. chief economist at UBS Securities LLC in New York who formerly worked at the New York Fed. “They have been ahead of the curve in looking at things from a different perspective, which adds a lot of value in the current environment.”

Directed by Atlanta Fed President Dennis Lockhart, a 69-year-old former Citigroup banker, and Research Director David Altig, reports introduced since 2009 have boosted the bank’s relevance in monetary policy-making.

The Atlanta Fed’s tracking of gross domestic product has proven more accurate since its introduction than consensus Wall Street forecasts, getting closer to the first report of U.S. growth than private estimates in four of the past six quarters. It estimated a 1.8 percent increase in the second quarter, closer to the actual 1.2 percent figure reported early Friday than the 2.5 percent median estimate of Wall Street economists.

Its inflation data has proven to be groundbreaking as well. The Atlanta Fed’s sticky-price index, which only measures prices that don’t move often, giving a signal on underlying inflation, has been followed by similar research by the Bank of England and the National Bank of Hungary.

Related: Report flags top 5 threats facing insurers

Atlanta’s approach changed in the wake of the 2007-09 recession, as economists sought to understand why the recovery was different than others, said Altig, research director who was hired by Lockhart from the Cleveland Fed.

The new tools are the product of the weekly 90-minute briefing by the top Atlanta economists each Tuesday for Lockhart and before major outlook speeches. Lockhart, who has never dissented on a Federal Open Market Committee policy decision, has sometimes been a bellwether of how the entire panel is leaning because he has been pragmatic and changed views with incoming data.

Fed held rates on Wednesday

“As the recovery progressed, we kept finding things.” Altig said. “The dynamics of the labor force, including the lack of wage growth, was the first thing to catch our attention. That inflation was for so long underneath the FOMC objectives.” The research focus “grew out of those sorts of questions.”

Many of the projects have been years in the works.

Altig originally asked economist Pat Higgins to put together the growth tracker, called GDPNow, starting in 2010. While Lockhart and his team regularly tracked consensus monthly forecasts by private economists, it was impossible to deconstruct exactly what was driving the often large changes, Higgins said.

Altig, Higgins and two other top researchers, Mike Bryan and Brent Meyer, all relocated from Cleveland. By 2011, Higgins had put together a statistical model that updated estimates of GDP based in part on work done by the Minneapolis Fed. The growth tracker was used internally for three years until it was released publicly in 2014.

Welcomed by many, the initiative has also drawn the attention of private forecasters who sell similar products. “Yes I am concerned that an alternative to our GDP tracking is being offered for free,” said Ben Herzon, an economist with Macroeconomic Advisers LLC in St. Louis, a company that markets its own popular GDP tracking forecast.

“But we get our tracking updates out sooner and more frequently, and we are always improving our methodology, which I believe will eventually translate into a demonstrably better track record,” he said.

The Atlanta Fed’s wage tracker, developed in 2014, was intended to give a more accurate indication of how typical workers are doing. The Labor Department’s report on average hourly earnings, which showed a 2.6 percent annual increase in June, represents an average of workers’ pay increases.

Related: Gross says negative rates are like ‘supernova’ that will explode  

Measuring wages

The wage tracker uses Census data to identify individuals and get their pay for a current month as well as what they reported a year earlier. Because average wage growth is reduced by retirement of higher-income earners and entrant of younger workers, the Atlanta Fed tally matches people’s actual experiences more closely among those with continuous job histories.

“We are beginning to see slightly faster wage growth,” Yellen said in congressional testimony June 22, citing the Atlanta Fed’s measure.

The wage tracker has shown a stronger correlation with measures of labor-market slack such as the unemployment rate.

“We saw what was happening with sluggish wages and asked, is there any other window we can get on wage dynamics,” said John Robertson, an economist and senior policy adviser. The report was developed based on research by the San Francisco Fed. UBS Securities, in a report July 12, cited the wage tracker showing “a more compelling picture of a tight labor market and inflation risk.”

The Atlanta Fed’s sticky-price inflation index has been showing the same trend as the wage tracker: pressures could be building.

Hungary introduced its own sticky-price index in 2013, citing the Atlanta research, noting the measures of underlying inflation “outperforms the core inflation measure” excluding food and energy prices. The Bank of England found the sticky index could help determine economic slack and “provide monetary policy makers with a steer on the medium-term inflation expectations of price-setters.”

Atlanta’s next research effort is a monthly survey of U.S. business people, now of 1,800 decision makers and intended to expand to 2,400, as a way to measure pricing trends. While most surveys ask consumers or households or economists about overall inflation expectations, this yet-to-be-released series asks what business people are actually doing with their prices.

Related: Insurance regulators respond to ongoing low interest rates