Skip to main content

The FRED® Blog

Paying interest on excess reserves An additional policy tool for the Fed

Commercial banks must adhere to regulations, including so-called reserve requirements. That is, banks must hold a certain fraction of their deposits as cash in a Federal Reserve account; these are known as “required reserves.” Banks can choose to hold even more cash in those accounts than what the Federal Reserve requires; these are known as “excess reserves.”

The graph above shows that required reserves are quite stable and grow as a constant fraction of total deposits in the banking system. But excess reserves increased considerably in 2008, as the Fed expanded the money supply to finance unconventional monetary policy measures such as quantitative easing. As of May 2018, excess reserves are nearly $1.9 trillion, ten times more than required reserves.

In normal times, excess reserves aren’t profitable, as they don’t earn a return. Instead of holding cash as excess reserves, banks could lend those funds and earn interest. However, after the 2008 recession, the Federal Reserve started paying interest on excess reserves (IOER). By altering the incentives for commercial banks to extend loans or hold excess reserves, the Fed is able to use the IOER as an additional monetary policy tool.

The second graph plots the IOER along with the (effective) federal funds rate, the Fed’s main tool for conventional monetary policy. The federal funds rate can be thought of as the interest rate at which financial institutions make short-term loans to each other. Here, we see that the federal funds rate tracks the IOER very closely. When banks have excess liquidity or reserves, they can choose whether to lend those reserves to other banks (at the federal funds rate) or deposit them at the Fed (and earn the IOER). Banks aren’t willing to lend to each other if the federal funds rate is substantially lower than the IOER, and so the two rates move closely together.

How these graphs were created: For the first graph, search for and select “required reserves of depository institutions” and click “Add to Graph.” From the “Edit Graph” panel, choose “Add Line,” search for and select the monthly “excess reserves of depository institutions” series, and click “Add data series.” The first series is in billions of dollars; to change it to match the second series (in millions of dollars), select “Edit Lines”/”Edit Line 1” and add the formula a*1000. For the second graph, search for and select the monthly “effective federal funds rate” series. From the “Edit Graph” panel, choose “Add Line” and search for and select “interest rate on excess reserves.” Use the date range tool to set the start date in August 2008.

Suggested by Asha Bharadwaj and Miguel Faria-e-Castro.

View on FRED, series used in this post: EXCSRESNW, FEDFUNDS, IOER, REQRESNS

Intermediate input dynamics Buying goods to make more goods

When a firm manufactures a good, the production uses not only labor and physical capital, but also intermediate goods and materials produced “upstream” in the production network. For example, when an automaker produces a car, it needs to purchase steel, glass, electronic devices, and more from other companies. The graph above shows the ratio of the costs of intermediate goods to total revenue (or gross output) for all U.S. industries. We see that the whole economy relies heavily on the production network, as the revenue share of intermediate inputs is above 40% on average. Also, the material share isn’t constant and actually fluctuates over time. During the Great Recession (2007-09), for example, the share drops from 46% to 41%, which implies that firms slowed down their purchasing of goods from each other. And, by 2011, they were buying and selling at about the same level as before the recession.

The second graph shows the material share specifically for manufacturing, which is an industry that uses more intermediate goods for production: Its average material share is 65%—much higher than the 43% for the entire economy. Its material share dropped as well in the recession, but in 2011-14 it overshot its pre-recession level, probably because firms were compensating for the amount of shipments they would have ordered (but did not) during 2008-09.

The final graph shows the FIRE sector (finance, insurance, and real estate). The FIRE sector typically uses less material (focusing on office equipment and delivery of services), but the material share is still above 30%. Interestingly, the level of material usage for the FIRE industry fell in 2008, but it has not yet recovered to its pre-recession level.

We’ve seen a solid rebound, an overshoot, and a shortfall in the material share soon after the recession. Overall, the numbers in 2015-17 suggest that a full recovery hasn’t yet occurred for the U.S. production network.

How these graphs were created: From the FRED homepage, select the option to “Browse Data by Release” below the main search bar. Choose “Gross Domestic Product by Industry” as the release. This release contains both intermediate inputs and gross output for all industries. Select the intermediate inputs by industry table and view the first table, in billions of dollars, seasonally adjusted at annual rates. Click on “Private Industries” to view the intermediate input cost of all private industries. To express as a share of gross output, use the “Edit Graph” menu to customize the graph: Add “Gross Output of All Industries” in the search box on the “Edit Line 1” tab. Type a/b in the formula box and click “Apply.” Repeat this process for each subindustry as desired.

Suggested by Sungki Hong.

View on FRED, series used in this post: GOAI, GOFIRL, GOMA, IIAI, IIFIRL, IIMA

The economics of oil sanctions A look at Iran, the law of one price, and the global bathtub

Recently, we’ve heard a lot about new sanctions the U.S. government may impose on the Iranian economy—in particular, against Iranian oil. Sanctions are a common policy tactic, but how do they work from an economic perspective?

First consider the supply of oil, which economists have described as a “global bathtub“: The tub is filled by “spigots” from various suppliers and depleted by “drains” from various consumers. The global oil price is determined by the sum of these supplies and demands for oil. The graph above shows global oil prices for West Texas Intermediate, Brent, and Dubai crude oil. We can see their global prices are fairly similar over time, with small differences between them. (Read this 2016 FRED Blog post for more info on these slight, temporary price differences between types of oil.)

The idea behind a global price for oil is the “law of one price” in international trade: If a homogenous good has negligible transportation or transaction costs, its price should be the same in all markets. This holds for crude oil, although not for a consumer good like a Big Mac, for example. Given that the global oil market is an integrated market, a shortfall in one region can be adjusted for by shipping the same or similar oil from another region in the world.

Now, what if a country is targeted by a ban on oil exports, as Iran was in 2012? The graph below show Iran’s oil production and oil exports, with noticeable declines beginning in 2012 that contributed to its 15-20% decline in per capita GDP. The last graph shows how the U.S. stopped importing oil from Iran, with the value of U.S. imports dropping to almost zero in 2012-15. Note: We can’t conclude from these data that the sanctions affected the price of Iranian oil, only that there was a decline in the quantity produced.

So why did these sanctions work in reducing oil exports from Iran? Economists say these sanctions were effective because of the international coalition that included key Asian countries that are heavy importers of oil. As for the global price of oil, it was fairly consistent in 2012-15, without any substantial changes. More FRED data series show how U.S. oil production has increased since 2012, compensating for the decline in production levels of other countries during that time. So, U.S. oil acted as a substitute for Iranian oil during this period and helped keep global oil prices stable. Exactly what we’d expect given the law of one price.

How these graphs were created: Search for “global price crude” (first graph) and “crude oil Iran” (second graph), select the series, and click “Add to Graph.” For the third graph, search for and select “goods imports Iran” and click “Add to Graph.”

Suggested by Suvy Qin and George Fortier.


Has the Canadian trade balance permanently shifted?

The U.S. trade balance, especially for goods, receives a lot of attention. Of course, FRED has data from all over the world, so we can stroll north a bit and look at the Canadian trade balance of goods.

Note the deep plunge in 2008, which Canada still hasn’t recovered from. First, it’s unusual to see a shock in the data this drastic and persistent, especially without a revolution or economic collapse of some kind. Second, for Canada, the Great Recession has looked much more like a normal recession than it has elsewhere in the world. So why the big chill in Canadian trade? Canada’s trade balance has always hinged on primary commodities. This country is a major exporter of all sorts of minerals and agricultural products—especially oil. These commodities are traded on world markets and are sometimes subject to wild price swings. And commodity prices did indeed run high until 2008, when they crashed. While they’ve recovered somewhat since then, the particular composition of Canadian imports and exports has kept the trade balance mostly negative. Is it going to stay that way? Check back in a few years, eh?

How this graph was created: Search for “Canada trade balance” and click “Add to Graph.”

Suggested by Christian Zimmermann.

View on FRED, series used in this post: BPBLTD01CAQ637S

The pitfalls of mapping poverty Definitions are important, but difficult to pin down

Where is poverty prevalent in the United States? There’s no simple answer to this question because notions of poverty differ. And defining an objective threshold for poverty is especially difficult. But we can use FRED to put forth a good-faith effort. The Census Bureau uses a specific procedure to categorize poverty and then provides the data: First, they categorize 48 types of American households—which vary by age and composition. Second, they determine what income counts toward the threshold. And third, they determine what that threshold is. They then estimate what proportion of these households live below that threshold.

The results at the county level are shown in the GeoFRED map above. Before interpreting it, though, one needs to keep in mind that there’s no geographical variation for the poverty thresholds. This means that two families with the same income may be considered to be living in poverty regardless of whether they live in a high-cost or low-cost county. This absence of regional adjustment may bias the map. Indeed, according to some (subjective) poverty standards, too many people may be counted as poor in Mississippi (where costs are rather low) and not enough may be counted in the Washington, DC, area (where costs are rather high). So, one should always be cautious when looking at poverty data.

How this map was created: Go to GeoFRED, select county-level data, and search for “Estimated Percent of People.” (This label isn’t very helpful, but the FRED Team is hoping to improve that in the near future.)

Suggested by Christian Zimmermann.

Subscribe to the FRED newsletter

Follow us

Twitter logo Google Plus logo Facebook logo YouTube logo LinkedIn logo
Back to Top