Federal Reserve Economic Data

The FRED® Blog

Understanding federal energy expenditures data

As part of its overall accounting of expenditures, the Bureau of Economic Analysis (BEA) collects data specifically on government spending, which can be broken down further into finer categories. FRED has time series of these data to help users explore the evolution of government spending. That evolution can be hard to interpret, so today we take a closer look at one of these series.

The FRED graph above tracks real federal energy expenditures from 1960 to 2022 using an index, which provides annual values relative to the level in 1976. Most of these expenditures come from activities by the Department of Energy (DOE).

During the 1960s and early 1970s, energy expenditures were much lower, ranging from around 10% to 40% of their 1976 level. Starting around 1974, energy expenditures climbed sharply, which coincided with the 1973 oil crisis and subsequent founding of the DOE in 1977.

Since the 1980s, energy expenditures have roughly kept pace with inflation but have varied widely year to year: Expenditures peaked at around twice their 1976 level in the early 1980s and again in 2010. By 2021,  real energy expenditures were back to about the same level as they were in 1976.

In 2022, energy expenditures actually become negative for the first time since reporting began in 1959. The FRED graph below zooms in on that time period and shows the year-over-year percent change in real energy expenditures since 2012. In a typical year, energy expenditures change by as much as 20%. But in 2022, energy expenditures declined by 115%. Why?

Just like the jump in spending during the 1970s, the drop in 2022 can be explained by policy responses related to events in global oil and gas markets.

The US stores a constant back-up supply of crude oil called the Strategic Petroleum Reserve (SPR). Within a typical year, very little of this store is released, but the US president can authorize drawdowns from the SPR during emergencies or shortages. For example, then-President Obama authorized a large release of oil from the SPR in response to the war in Libya in 2011. After Russia’s 2022 invasion of Ukraine disrupted oil markets, President Biden authorized a large “emergency drawdown” of the SPR in coordination with the International Energy Agency’s International Energy Program.

Over the course of 2022, this drawdown added around 180 million barrels. For context, the last two emergency drawdowns in 2011 and 2005 were around 30 and 20 million barrels, respectively. (For more discussion of petroleum reserves, check this May 2024 FRED Blog post.)

When the government releases oil from the SPR, it does so through competitive sales, which are accounted for in current expenditures as net outlays and are negative values. In 2022, SPR-related outlays were large enough to dwarf other energy-related spending, resulting in negative total expenditures.

How these graphs were created: First graph: Search FRED for series “G160551A027NBEA.” Using the date selector above the graph, set the date range to 1960-01-01 to 2022-01-01. From the “Edit Graph” panel under the “Edit Line 1” tab, use the “Customize Data” section to search for and select the series “CPIAUCSL” to add to the graph. Apply the formula a/b. Under “units” at the bottom, convert your new combined series to an index and set 1976-01-01 as the custom date. Second graph: Switch the final units of the first graph from an index to percent change from a year ago. Use the date selector to change the date range to 2012-01-01 to 2022-01-01.

Suggested by Bill Dupor and Marie Hogan.

Signals of continued economic resilience from the output gap

At the July 31 FOMC press conference, Chair Powell said “recent indicators suggest that economic activity has continued to expand at a solid pace,” even in the face of a labor market that appears to be normalizing.

Real GDP growth picked up significantly in the second quarter of 2024, according to the BEA’s advanced estimate, at an annualized rate of 2.8%. While this output growth is an important measure of activity on its own, many policymakers also pay attention to actual output growth with respect to potential output growth, the economy’s estimated maximum sustainable output.

The FRED graph above shows the output gap, which is the difference between actual and potential real GDP. Output has been above potential for the past year; most recently, it appears to have exceeded potential output by about 1% in the second quarter, up from approximately 0.9% in the first quarter. This is an improvement over 2022 and early 2023, when the output gap was slightly negative, and is roughly in line with the second half of 2019, when economic growth was relatively high compared with its 2010-2018 average.

The graph also shows the output gap tends to be negative after recessions, but then eventually returns to a positive gap after a recession. For example, the 2008-2009 financial crisis was deep and long enough to keep actual output below potential all the way through 2017. It wasn’t until late 2019 that the gap became firmly positive around levels not seen since 2007.

One signal of the economy’s resilience is that output returned to potential within two years of the initial COVID shock in 2020 despite a precipitous decline. Another signal is that the output gap has continued to become positive even as potential output has continued to grow at a stable pace: Annualized quarterly growth rates of potential output have hovered around 2% since 2018. By comparison, growth rates of GDP have averaged 2.8% over the past eight quarters. This also has implications for monetary policy, referred to in this FRED Blog post about the Taylor Rule.

An important caveat: Potential output cannot be observed, so policymakers contend with considerable uncertainty here, including frequent and unpredictable revisions to the estimate of potential output. A good illustration of these revisions over time was delivered by Larry Summers in the 2016 Homer Jones Memorial Lecture.

How this graph was created: In FRED, search for and select “Real Potential Gross Domestic Product.” From the “Edit Graph” panel, use the “Customize data” section in the “Edit Line 1” tab to search for and select “Real Gross Domestic Product.” You should see two series on the “Edit Line 1” tab listed as (a) and (b). In the “Customize data” section again, enter and apply (b/a – 1) * 100 in the formula bar.

Suggested by Kevin Kliesen and Joseph Martorana.

Trends and cycles in US productivity

Today we look at total factor productivity, which is a measure of economic efficiency that captures how effectively an economy uses its inputs to produce outputs. TFP reflects technological progress through innovation and adoption of new technologies, allocation of resources, and other factors that boost the overall economic product beyond just increases in labor and capital.

Improvement in TFP is crucial because it drives long-term economic growth and raises living standards. The FRED graph above shows two interesting observations related to this.

First, over the long run, US TFP has grown significantly: Between 1955 and 2015, it improved by about 55%. This increase represents a substantial improvement in the nation’s ability to generate economic output from its resources.

However, the rate of TFP growth has slowed noticeably in recent decades. Particularly since 2005, TFP has increased by only about 5%. This slowdown is a concern for economists and policymakers because it suggests a potential decline in the pace of innovation or the economy’s ability to adopt new technologies.

Second, the graph also clearly shows that TFP tends to significantly drop during recessions, indicated by the shaded areas. This doesn’t necessarily mean that the economy literally “forgets” how to produce goods and services efficiently. Instead, these dips reflect several economic realities during downturns: For example, capacity utilization often decreases, leading to less-efficient use of existing resources. As the economy recovers, TFP typically rebounds, suggesting that these efficiency losses are generally temporary rather than permanent losses of productive knowledge or capability.

How this graph was created: Search FRED for “total factor productivity” and click on the series for the United States.

Suggested by Aakash Kalyani.



Subscribe to the FRED newsletter


Follow us

Back to Top