Federal Reserve Economic Data

The FRED® Blog

Negative nominal interest rates for real?

A true story in Switzerland

Real negative interest rates are easy to imagine when inflation is higher than the interest rate. But nominal negative interest rates have long been thought of as either inconceivable or unsustainable. Yet, in recent years, several European countries and Japan have made negative nominal interest rates a reality. The most extreme case seems to be Switzerland, which is featured in the top graph: The spot rate, the 3-month LIBOR, and even the 10-year government bond rate are all negative now and have been for several years. How is this possible?

This isn’t a case of an economy that needs major stimulus through low interest rates. Rather, it’s an export-focused economy whose currency has a strong tendency to appreciate; in fact, the Swiss franc is considered a refuge currency in times of crisis. The crisis at hand involves the euro’s various troubles in recent years, including the debt problems of some of its member countries. Switzerland has avoided these troubles and has even managed to achieve successive government surpluses. So it’s easy to understand why there’s so much demand for Swiss francs and bonds. But so much demand typically causes a local currency to appreciate, which would make exporting more difficult. The Swiss National Bank, therefore, has adopted a policy of negative interest rates to make the franc less attractive. Interestingly, the effect permeates the Eurodollar market as well, as shown in the bottom graph.

How these graphs were created: NOTE: Data series used in these graphs have been removed from the FRED database, so the instructions for creating the graphs are no longer valid. The graphs were also changed to static images.

Suggested by Christian Zimmermann.

View on FRED, series used in this post: CHF3MTD156N, IR3TED01CHM156N, IRLTLT01CHM156N, IRSTCI01CHM156N

The education trade balance

Our last post discussed the foreign travel trade balance—that is, money spent by foreigners for tourism, business, and education in the U.S. compared with what Americans spend abroad. In this post, we focus on travel for education. The Bureau of Economic Analysis began including travel-for-education data only recently, when they made their last big revision to GDP data. In fact, the BEA has recomputed these data for the past several years, so we can compare the old and new series to get a sense of the education trade balance. (This is a much more convenient process than adding up all the expenses of foreign students, so thanks BEA!)

The FRED graph above makes it quite apparent that foreign students spend much more in the U.S. (3.5 times more) than American students spend abroad. In fact, the surplus for the last month of available data, $3.3 billion in September 2017, is about 7% of the overall trade deficit at the time and 12% of the trade deficit with China. Obviously, this surplus is quite substantial—with the bonus of having top international students study in the U.S., and possibly remain as part of the U.S. economy, after having their education paid by funds from abroad.

How this graph was created: Search for “travel trade,” select the exports series that includes education travel, and click “Add to Graph.” From the “Edit Graph” panel, add the export series that removes education travel and apply formula a-b. Repeat for the imports series.

Suggested by Christian Zimmermann.

View on FRED, series used in this post: BOMTVLM133S, BOXTVLM133S, ITMTAEM133S, ITXTAEM133S

The business behind the trade balance

Why trade deficits decrease in recessions and increase in booms

How does the trade balance relate to economic activity? The graph above shows the U.S. trade balance for goods and services as a percentage of GDP. Obviously, there was a surplus initially and now there’s a persistent deficit. Beyond that, it looks like every time there’s a recession, the trade deficit tends to decrease. (Or, if we go farther back in the past, the trade surplus tends to increase.) Obviously, many things affect the trade balance, but let’s see what FRED can show us about this relationship.

A good way to reveal how series may be correlated is to look at scatter plots. Instead of relating economic data to dates, scatter plots relate two data series to each other, one on each axis. The graph above does this with changes to the trade balance ratio on one axis and percent changes to real GDP on the other axis. What may look like a random assortment of dots actually has some information. Imagine the graph is divided into four quadrants and then consider where the dots are located. The upper right and lower left quadrants have fewer data points than the other two, highlighting that there is indeed a negative correlation: That is, when real GDP tends to increase, the trade balance tends to decline—that is, trade surpluses decrease or trade deficits increase.

Why is that? First, consider that the trade balance is net exports—that is, exports minus imports. Imports are highly correlated with GDP, while exports are less so. We see this in the graph above, which plots imports. This time, the upper left and lower right quadrants are the most populated. This highlights the positive correlation: That is, when real GDP tends to increase, imports do as well. Thus, over the business cycle, it is really imports that drive the trade balance: When the economy is doing well, producers need more intermediate goods, and imports are mostly intermediate goods. Also households consume more, and a share of those consumption goods are imports. If you graph exports, the correlation is much harder to see. Exports depend much more on what happens abroad, which isn’t that well correlated with domestic activity.

How these graphs were created: First graph: Search for “net exports” and select the quarterly series. From the “Edit Graph” panle, add GDP and apply formula a/b*100. Second graph: Use the first graph and change the sample period to start in 1954. From the “Edit Graph” panel, change the units to “Change.” Add a line by searching for “real GDP,” change its units to “Percent change,” open the “Format” tab, and switch the type to “Scatter.” Third graph: Use the second graph but with real imports in percent change.

Suggested by Christian Zimmermann.

View on FRED, series used in this post: GDP, GDPC1, IMPGS, NETEXP


Back to Top