Today’s post explains the relationship between interest rates and exchange rates and how they’re involved in investment decisions.

The data

The FRED graph above tracks two rates:

• The solid red line shows the exchange rate between the US dollar and euro.
• The dashed blue line shows the difference in interest rates (or yields) between the long-term/10-year US Treasury bond and the German government bond.

We can see in the graph that these rates appear to be related: When this interest rate differential (US bond yield minus German bond yield) has increased, the US dollar has tended to appreciate.

But exchange rates are affected by many factors, especially shocks that alter market views about the long-run future exchange rate. Such a shock appears to have occurred after April 2, 2025 (a.k.a., “Liberation Day”).

The graph shows that, at the time, US Treasury yields rose sharply relative to German government bond yields. In theory, that would have implied a stronger US dollar. But instead, the dollar depreciated. Market participants seem to have revised down their expectations of the dollar’s long-run value, possibly due to concerns that large tariffs would erode US economic fundamentals.*

For a deeper look, read on…

Investing decisions

Economists see a tight link between these interest rate differentials across countries and the expected changes in the exchange rates. So, in theory, investing in domestic bonds or foreign bonds should yield roughly the same rate of return. Again, when the interest rate differential (US bond minus German bond) increases, the US dollar tends to appreciate. This pattern supports the validity of the economic concept known as uncovered interest parity, or UIP. UIP states that

Domestic interest rate ≈ Foreign interest rate + Expected depreciation of the foreign currency

or

Expected change in the exchange rate ≈ Interest rate difference between the home and foreign countries

Bond yields in both the home and foreign countries are known at the time of investment and therefore they involve little uncertainty. But exchange rates are another matter.

Investors in foreign bonds must first convert their funds into the foreign currency to buy those foreign bonds. Then they must convert their funds back to the domestic currency once the foreign bond matures. The future exchange rate isn’t known at the time of investment, so the return from investing abroad is uncertain because of that exchange rate risk.

Long-run theory versus short-run observations

These patterns still support the validity of a long-run UIP through a mechanism in which today’s exchange rate adjusts to maintain parity, rather than through a shift in the long-run average exchange rate. But empirical evidence shows that, in the short run, exchange rates are nearly unpredictable and behave close to a random walk. This suggests that UIP doesn’t hold in the short run.

The data support UIP much better in the long run. Over longer horizons, exchange rates tend to revert to the mean. When there’s an increase in the interest rate differential between the home and foreign countries, today’s exchange rate should appreciate immediately in the higher-interest-rate country so that the expected exchange rate depreciates in the future as it swings back toward its long-run average. In other words, long-run UIP can hold through an exchange rate adjustment occurring today, rather than through changes in the long-run average exchange rate.

Consider this example: Suppose the long-term domestic interest rate remains unchanged at 5%, while the foreign long-term interest rate suddenly falls from 4% to 3%. Long-run UIP implies that investing abroad should still yield roughly a 5% return on average, despite the lower foreign interest rate. The 1- percentage-point reduction in the foreign interest rate should therefore be compensated by an expected appreciation of the foreign currency of about 1% annually. To achieve this, the domestic currency must appreciate immediately, so that it is subsequently expected to depreciate as the exchange rate converges back to its long-run average.

*See Jiang et al. (2025).

How this graph was created: Search FRED for and select the US Dollars to Euro Spot Exchange Rate (DEXUSEU) series. Click on “Edit Graph”: Under the “Edit Line” tab, modify the frequency to monthly and scroll down to the formula box and enter 1/a. Thus, instead of a USD/EURO exchange rate we have a EURO/USD exchange rate. Next click on “Add Line” and enter DGS10 to add the yield for 10-year US government bonds. Modify the frequency to monthly. Under “Customize data,” add the series IRLTLT01DEM156N, which is the 10-year German government treasury bond yields. Change the formula to a-b. From the “Format” tab, under line 1, click customize and change the color to red. Under line 2, click customize and change the color to blue and place the axis on the right. Finally, edit the dates so that the series starts on January 1, 2024.

Suggested by YiLi Chien and Kevin Bloodworth.

The title of this post may have reminded you that you need to buy some chocolate for some event in a couple of days. If you do, you may also notice that chocolate has become quite expensive. If you already made the trip, you may have bought less than usual or switched to some other sweet product. Either way, let’s look at the price of chocolate.

First, let’s be clear that chocolate has indeed become more expensive. Our FRED graph above shows the evolution of two types of candy: those with cacao-based chocolate and those without. The prices of both types have increased lately, but it’s very clear that chocolate and its derivatives have become significantly more expensive.

Why? The main ingredient of chocolate is cacao. (Cocoa is the term for its roasted form.) Its cultivation is concentrated in a few countries for climatic reasons, and it’s not produced domestically in the US. Cacao crops have been particularly bad in the past couple of years.

• Because of climate changes, current cacao trees aren’t optimal for their location.
• New trees take a while to grow and take 3 to 4 years to bear fruit.
• A virus is afflicting current plantations.

This lack of cacao supply has led to a marked increase in the world price for this commodity, as seen in our FRED graph below.

From a US perspective, do tariffs enter into the picture? The US imposed “reciprocal” tariffs on cacao-producing countries in February 2025, typically 15%. But these tariffs and some for other commodities that cannot be grown in the US were removed in November 2025. Thus, tariffs shouldn’t be a factor for this year’s Valentine purchase unless your purchase isn’t that fresh.

How these graphs were created: Search FRED for “Chocolate products” and select the right series. Click on “Edit Graph” then on the “Add Line” tab. Search for non-chocolate and select the right series. Click on the “Edit Lines” tab and select units “Index (Scale value to 100…)” with date 2011-12-01 and click on “Copy to all.” Open the “Format” tab, change the color of the first line to brown and the frame to pink. Finally start the graph on 2011-12-01. For the second graph, just search for “cocoa world price.”

Suggested by Christian Zimmermann.

On February 5, 2026, the Bureau of Economic Analysis released their 2024 real GDP breakdown at the county level. Here are some highlights from the data set, some of which are shown in the FRED map above:

• In 2024, real GDP growth was positive in three-quarters of all counties.
• Nationally, real GDP increased by 2.8%. However, the median county experienced growth of 2.3%. About two-thirds of counties experienced growth ranging from -1.6% to 6.0%.
• The county with the fastest growth was Carter County, Montana, at 76.6%.
• The county with sharpest decline was Baca County, Colorado, at -46.3%.
• There was a positive relationship between real GDP growth and the size of the county. Among the largest 10% of counties, growth averaged 3%; whereas, among the smallest 10% of counties, growth averaged -1.5%.
• The county with the fastest growth here in the St. Louis, Missouri-Illinois metro area was Madison County, Illinois, with 8.3% growth. Jersey County, Illinois experienced the slowest growth in the metro area, at -1.2%.

As noted above, there are some large numbers for growth and contraction of real GDP at the county level. This is because many counties are very small. Therefore, GDP can fluctuate greatly from one year to the next: Economic shocks such as a business openings or closings in a small town can have a significant impact on the community and, thus, the economic data. There are many reasons why some counties grow while others contract. For example, the industrial composition can amplify the degree of expansion or contraction in relation to the national overall business cycle. Demographic makeup and migration patterns of a county can also be a factor. These reasons are explored in more detail in this St. Louis Fed essay.

How this map was created: Search FRED for and select “Real GDP County” and click on the first choice. Click on the “View Map” and then “Edit Map” buttons. Change units to “Percent Change from Year Ago.” Then switch the number of color groups to 3 and “data grouped by” to “User Defined Method”; then define the scales at 0, 3, and the highest value (which is 77). For values less than 0, choose red to show contraction; for values less than 3, choose light green to show slow to moderate expansion; for values less than 77, choose dark green to show rapid expansion.

Suggested by John Fuller and Charles Gascon.