Federal Reserve Economic Data

The FRED® Blog

Just the facts, ma’am

Embracing seasonality in national accounts data

Usually, if you have the choice, you want to look at macroeconomic data that have been seasonally adjusted. This adjustment lets you compare periods within any year without being misled by the various fluctuations that occur every year. Ice cream production, tourism, and toy purchases, for example, all have predictable seasonal factors, and usually we’re interested in what happens beyond these seasonal effects.

The same logic applies to aggregate measures such as gross domestic product (GDP). But, occasionally, we just want the raw data. All three GDP components shown in the graph above consistently dip in the first quarter. In most of the country, the winter months hinder activities such as residential construction and local government infrastructure projects. As for the dip in imports, that’s linked to the dips described above, because general declines in activity are definitely reflected in imports.

How this graph was created: From the gross domestic product release, go to section 8 (not seasonally adjusted); then choose the table with real GDP, select the relevant series, and click “Add to Graph.”

Suggested by Christian Zimmermann.

View on FRED, series used in this post: ND000338Q, ND000343Q, ND000351Q

The return of swap rates

Tracking interest rate risk in FRED

First, some background on swaps: Let’s say you’re borrowing at an adjustable interest rate that fluctuates along with the LIBOR (London interbank offered rate). Let’s also say you’re not so comfortable with the LIBOR’s fluctuations. You can engage in a swap and get someone else to pay that fluctuating interest rate for you, while you pay them a constant interest rate. The constant rate you pay is the swap rate. Now the swap rate stays constant during the lifetime of each individual contract, but the swap rate you can expect to pay for a new contract changes all the time; in fact, many swap rate series in FRED are updated daily at various times.

So what’s interesting about all this? The graph shows the 12-month swap rate, which combines the old (red line) and new (blue line) sources of the data,* and the LIBOR. Note that the swap rate and the LIBOR are different. This difference has to do with expectations of the future evolution of the LIBOR, risks attached to the LIBOR and the swap contract, and the implicit insurance that you get from a swap. Notice that the difference increased pretty significantly at the time of the financial crisis; it has narrowed recently but is still noticeable. If interest rates have stayed quite low, what’s behind these changes?

There are at least three possible factors: First, a swap carries the additional risk that your counterparty may default, in which case you’ll be on the hook for paying the interest. That risk of default may have increased—or at least the perceived risk may have increased. Second, there have been doubts in recent years that the LIBOR may not accurately or transparently reflect market rates. Third, expectations about the evolution of the LIBOR during the length of the swap contract may have trended away from the current value more than they had in the past.

*NOTE: FRED had included swap rate data from the Board of Governors of the Federal Reserve System until the series were discontinued in 2016. FRED has since replaced—swapped?—those important financial data for data now supplied by IBA (Intercontinental Exchange Benchmark Administration).

How this graph was created: NOTE: Data series used in this graph have been removed from the FRED database, so the instructions for creating the graph are no longer valid. The graph was also changed to a static image.

Suggested by Christian Zimmermann.

View on FRED, series used in this post: DSWP1, ICERATES1100USD1Y, USD12MD156N

Hidden figures in homeownership

New homeownership rates by race and region

FRED recently added U.S. Census Bureau data on homeownership rates by race. The rates for each group, shown in the top graph, basically reflect the groups’ positions in the wealth distribution.

Changes in the homeownership rate are driven by small seasonal factors and longer trends. There was a decrease for all groups after 2006, which could reasonably be attributed to the Financial Crisis. Rates rebounded soon after for all groups except African Americans, whose rate has trended down pretty steadily since then.

This last fact could be due to regional differences: That is, homeownership rates could be decreasing in certain geographical areas where more African Americans live. The graph below shows very large sections of the country (Census regions), but doesn’t indicate any steady declines in any areas, including the South. So something else is at work that simply isn’t visible here.

How these graphs were created: On the Housing and Homeownership release table, select the series you want displayed and click “Add to Graph.”

Suggested by Christian Zimmermann.

View on FRED, series used in this post: ANHPIHORUSQ156N, AORHORUSQ156N, BOAAAHORUSQ156N, HOLHORUSQ156N, NHWAHORUSQ156N, RHORMWQ156N, RHORNEQ156N, RHORSOQ156N, RHORWEQ156N


Back to Top