Federal Reserve Economic Data

How and why do financial markets react to war? One aspect of war is to endure losses, and financial markets typically don’t respond enthusiastically to even the risk of loss, let alone widespread destruction on their own soil. Markets may also rise and fall over the course of the war as the fortunes of the warring parties change.

FRED has a peculiarly helpful dataset that provides examples of this dynamic: Weekly U.S. and State Bond Prices, 1855-1865. The authors, Gerald P. Dwyer Jr., R. W. Hafer, and Warren E. Weber, compiled a time series of bond prices for some U.S. states leading up to and through the U.S. Civil War.

The FRED graph above shows bond price data for two states in the South (Virginia and Louisiana) and two states in the North (Pennsylvania and Ohio).

The bonds in question were used for infrastructure, such as roads, canals, and railroads. The war’s potential to destroy that infrastructure could affect the ability of states to pay back the bonds. More generally, war disrupts productive capacity, impedes the raising of tax revenue, and ramps up state expenses—all of which increases the likelihood of state default.

Not surprisingly, the graph shows increasing risk, with prices dropping in late 1860 and then plunging as hostilities began in 1861. These effects were more pronounced in the South than in the North. And the graph also shows that Southern bonds stayed low as the war unraveled, while Northern bonds roughly returned to parity.

How this graph was created: Search FRED for “disunion bonds” and click on, say, Virginia. From the “Edit Graph” panel, use the “Add Line” tab to search for and select the other states.

Suggested by Christian Zimmermann.

Obviously, major wars take their toll on a country’s population. They also affect economies in distinct ways. For example, wars affect the manufacturing sector as firms ramp up production of military vehicles, munitions, and the like. The current war in Ukraine, while far away from the United States, may still be having an impact here, given that the United States has promised military equipment to Ukraine. Other countries have done the same and are also ramping up their own purchases. FRED has some related data (at least back to 1994) that may help show what’s happening on the manufacturing front.

The Manufacturers’ Shipments, Inventories, & Orders survey from the U.S. Census Bureau doesn’t detail the defense sector, but it does provide data on manufacturing with and without defense. So, we can graph the difference.

The first graph shows new orders. If the war was a complete surprise and governments are only now scrambling to acquire military equipment, we’d expect new orders to be significantly up. At the time of this writing, that does not appear to be the case. But maybe they had enough foresight and are taking deliveries now. The second graph looks at shipments. While there’s an increase, it appears to follow a trend that predates the Ukraine war quite a bit.

Another impact could be that the new demand for armaments is reducing manufacturers’ inventories. Our last graph looks at this, and inventories are actually up. Could this be in anticipation of increased demand in the near future? We can’t tell simply by looking at the graphs. So, in conclusion, we don’t see any hard evidence that this war has had any notable effect on U.S. manufacturing yet.

How these graphs were created: For each graph, start by searching FRED for the series (say, manufacturers inventories) adding the “defense” keyword to narrow the results. Once you have the graph, click on “Edit Graph,” add the other series by searching for the same keywords without “defense,” and apply formula b-a.

Suggested by Christian Zimmermann.

The FRED map above shows the proportion of disconnected youth in each U.S. county in 2020. Disconnected youth are not teenagers who’ve been grounded without internet privileges. They’re between the ages of 16 and 19, they’re not enrolled in school, and they’re unemployed or not in the labor force. These youth are typically thought to be at risk of future low income or even crime. So it’s important to try to learn something about them, such as how many there are and where they are.

Darker greens show larger proportions, but not necessarily larger concentrations. Remember that the size of a county is reflective of its land mass, not necessarily its population. In fact, large counties are typically the least-populated ones. For this reason it can be difficult to draw big conclusions from a map, but we can find some insights by comparing maps. Below, we show the same map with data from 11 years ago. (An earlier post compared similar data from 2010 and 2015.)

What’s remarkable when comparing these two maps is that they look quite alike. Darker counties in one are typically also darker in the other. Understand that the two maps do not look at the same people. The 16- to 19-year-olds in 2009 are 27- to 30-year-olds in 2020, and thus not part of the statistic. (Strictly speaking, this calculation is not that simple, as the statistics draw on 5-year averages, but the separation of cohorts still holds). This shows that there is high regional persistence for this kind of issue.

How these maps were created: Search FRED for “disconnected youth,” click on any county data, then click on the map button. Change years using the date picker.

Suggested by Christian Zimmermann.