March 29, 2012
Updated Estimates of Gross State Product Growth for the Seventh District
by Scott Brave and Norman Wang
This blog serves to expand on our December 2011 Chicago Fed Letter (CFL) by further detailing the estimation process used to produce estimates of annual Gross State Product (GSP) growth on a quarterly basis for the five Seventh Federal Reserve District states. In addition, we preview the estimates of GSP growth for 2011 that will be included as part of tomorrow’s Midwest Economy Index (MEI) release.
Last year, the Chicago Fed began releasing the MEI, a weighted average of 134 state and regional indicators that measures growth in nonfarm business activity. Two separate index values are constructed, the MEI (absolute value), which captures both national and regional factors driving Midwest economic growth, and the relative MEI (relative value), which provides a picture of the Midwest’s economic conditions relative to the nation’s.
MEI values correspond to deviations of growth in Midwest economic activity around its historical trend. Values above zero indicate growth above its historical trend, and values below zero indicate growth below trend. For the relative MEI, a positive value indicates that regional growth is further above its trend than would typically be suggested based on the current deviation of national growth from its trend, while a negative value indicates the opposite.
Together, the MEI and relative MEI provide a picture of the Seventh District’s state economies that is closer to being in real time than does the BEA’s GSP data. By exploiting the historical correlation between GSP growth in each of the five states and the MEI, we are able to produce quarterly estimates of GSP growth ahead of the annual BEA release of GSP data.
The statistical model we use to explain the annual growth in GSP for each Seventh District state is as follows:
The model succinctly summarizes the historical relationships between national (real GDP growth), regional (MEI and Relative MEI), and state-specific (lagged GSP and state real Personal Income growth) factors driving each Seventh District state’s GSP growth since 1979.
State-specific growth factors dominate in explaining Indiana’s, Iowa’s, and Michigan’s GSP growth, while national factors dominate in explaining Illinois’s and Wisconsin’s. Regional growth factors, on the other hand, vary in importance from 11% in Michigan to 39% in Wisconsin and are above 20% of the explained variance for Illinois, Indiana, and Wisconsin.
The regression coefficients estimated for our model are listed in the table below. Each coefficient represents the “effect” of each input on GSP growth. For example, a 1% increase in GDP growth leads to about a 0.5% increase in GSP growth across the Seventh District states, with the effect slightly higher for Illinois and Iowa and slightly lower for Indiana and Michigan (second row).
By plugging the latest data for GDP, MEI and relative MEI, state Personal Income, and GSP into the above equation, we can use the regression coefficients above to obtain a GSP growth projection for the current year. The remainder of the blog details how this process works in practice.
To make our out-of-sample predictions of GSP growth using the above model, we need current year values for all the inputs in our regression. Lagged GSP growth is available, but in the first three quarters of a year quarterly GDP growth, state Personal Income, and the monthly MEI and relative MEI only cover part of the year.
To obtain the annual growth rate in state Personal Income and national GDP in the first three quarters of the year, we average the quarterly values available in the current year and take the log first difference from the quarterly average of the prior year. In this respect, once every quarter we are able to make a prediction of annual GSP growth for each state based on year-to-date growth in these measures.
The MEI and relative MEI predictions are similarly constructed using the March, June, September, and December MEI values. Since the MEI and relative MEI represent three-month moving averages, the March MEI number captures the first quarter of activity, the average of the March and June MEI numbers captures the first two quarters of activity, and so forth.
Shown below are the regional (left-hand scale) and national (right-hand scale) growth factors described above. Both the MEI and relative MEI began the year nearly one standard deviation above their historical averages, suggesting that the Midwest Economy experienced rates of growth that were both above-average and higher-than-normal given the level of national growth. During this same period GDP growth was very weak; but over the course of the year, the national economy strengthened while the Midwest economy expanded at a slower rate.
Coming into 2011, there was considerable variation in state-specific growth factors with Indiana standing out as having by far the highest GSP growth rate in 2010 among the five Seventh District states and with Illinois having the lowest. In 2011, however, the state Personal Income data suggest Iowa experienced stronger growth than the other four District states, and Indiana and Illinois were instead clustered closely together with the remaining two District states.
Projections for annual GSP growth made through the second quarter, third quarter, and for all of 2011 in each of the five Seventh District states are displayed below. The growth projections for Iowa, Indiana, Michigan, and Wisconsin exceeded national GDP growth in 2011, while Illinois is projected to be slightly below the national growth rate.
The diversity we see across states is a direct consequence of the results for national, regional and state-specific growth factors mentioned earlier. For instance, Illinois’ relative weakness among the five states stems primarily from modest GDP growth in 2011, on which its forecast heavily depends. Weaker national growth is also responsible for the lower rate of GSP growth for Wisconsin in 2011.
On the other hand, the lower rate of GSP growth for Michigan in 2011 can be traced back to expected mean reversion offsetting the positive contributions of the MEI and personal income growth. Strong regional and state-specific growth factors boost GSP growth in Indiana above Illinois and Michigan, whereas state-specific growth factors, particularly high personal income growth in 2011, keep Iowa’s GSP growth rate steady from 2010 and much higher than the other states in the District.
The variation over the course of the year in our forecasts is also informative. Illinois’ GSP growth forecast strengthened throughout the year as national GDP growth increased. Wisconsin’s GSP growth forecast was also strongly influenced by national factors; but being more affected by regional factors than Illinois, increased only slightly over the course of the year. GSP forecasts for Iowa, Indiana, and Michigan all rebounded in the fourth quarter after weakening in the third quarter, closely mirroring the pattern of the Personal Income data for each state.
Our quarterly estimates of GSP growth can be found as part of the press release for the MEI following the third release of national GDP data for each quarter. The 2012 release schedule for the MEI can be found at www.chicagofed.org/mei.
March 28, 2012
Trends in Motor Vehicle Trade—A U.S. Perspective
by Thomas Klier
Motor vehicles tend to be sold near where they are produced. However, when local demand does not suffice to support a dedicated assembly plant, some vehicles are shipped across longer distances, including across oceans. The share of newly produced light vehicles (cars and minivans, sports utility vehicles or SUVs, and pickup trucks) exported from the U.S. to countries other than Canada or Mexico averaged only 4% of U.S. light vehicle production between 1996 and 2011. The share of new vehicles imported from outside the NAFTA (North American Free Trade Agreement) area was somewhat higher over this period, at 19%, as the U.S. has traditionally run a trade deficit in cars and light trucks.
This blog updates our earlier analysis on U.S. vehicle exports and adds some discussion related to vehicle imports.
During the past two years, exports of new and used vehicles from the U.S. have continued the strong growth exhibited since 2003. After the sharp decline of activity during the latest recession, exports rose again sharply and, by the end of 2011, had nearly matched the pre-recession peak reached in 2007 for new vehicles and in 2008 for used ones (figure 1).
Table 1 provides more detail regarding destination countries and regions for exports of new and used light vehicles from the U.S. While Canada and Mexico together represent the primary destinations, the relative importance of these two NAFTA partners has fallen a bit since 2007 (representing 55% in 2011, down from 64% four years ago), as most of the growth in exports of U.S. produced new vehicles originated from elsewhere (figure 2). Last year, Germany and China were as important as Mexico as destination for U.S produced new vehicles. Exports of used vehicles, on the other hand, are much more dispersed; Nigeria, Benin, and the United Arab Emirates jointly account for nearly one-third of all U.S. used-vehicle exports.
One of the factors behind this trend is the implementation of new trade agreements, such as the U.S. free trade agreement with South Korea, which was ratified in November 2011 and reduces South Korea’s tariff on passenger vehicle imports from 8% to 4%. Both the Detroit-based carmakers (Ford, General Motors, and Chrysler) and Honda and Toyota have announced a noticeable increase in exports of vehicles produced in the U.S. to South Korea. In the case of the Japanese-based carmakers, the decision to export to South Korea from North America instead of from Japan is likely influenced by exchange rate trends, which have strengthened the yen for some time.
Another phenomenon that has coincided with the increase in the share of non-NAFTA destinations has been the arrival in the U.S. of premium producers, such as Mercedes and BMW, that now ship output from their U.S. locations to countries around the world. While the U.S. and more generally, North America, very likely represent the largest single market for vehicles produced at the U.S. plants of these two producers, they tend to export a much higher share of their the U.S.-based production than a typical mass producer of vehicles. Furthermore, their U.S.-based plants are the sole producers of specific products, suggesting that their role in motor vehicle exports from the U.S. is not being jeopardized by the current lower rates of capacity utilization in the European motor vehicle industry.
Data on imports of vehicles sold in the U.S. are available for a much longer time frame. Vehicle imports have averaged 20.6% of U.S. sales since 1980. That share declined rapidly during the decade between the mid-1980s and the mid-1990s—a time when foreign-based producers quickly expanded their production capacity within North America, ramping down their imports of finished vehicles from overseas in turn. This “onshoring” of vehicle production has a corresponding impact on the trade in motor vehicle parts. When an overseas producer’s vehicle assembly plant is first set up in the U.S., the so-called domestic content, that is, the share of parts sourced from within the U.S., is noticeably lower than that of a comparable native assembly plant. Over time, however, the domestic parts content of vehicles produced by foreign-headquartered producers tends to rise significantly.
Since the mid-1990s, the import share of U.S. sales has been trending up again as additional companies have entered the U.S. market. The small car segment, of heightened interest to consumers in times of rising gas prices, represents a relatively small share of the overall market in the U.S. and many of the small vehicles sold in the U.S. tend to be produced overseas. However, the import share of U.S. vehicle sales has declined by 4 percentage points during the past two years (figure 3) despite rising gas prices --gas prices were rising from the beginning of 2009 through May of 2011. Figure 3 also illustrates that cars continue to represent a large majority of imported vehicles.
Figure 4 shows that the decline in the vehicle import share results from a decline in imports from Asia, as European imports have continued their steady increase exhibited over the last 15 years. Japan, by far the largest source country of U.S. vehicle imports from Asia, was negatively affected last year by the earthquake and subsequent tsunami. However, the decline in imports starts earlier than that, suggesting a contributing role of the yen/dollar exchange rate, which started a steady decline (strengthening of the yen) in 2007. In fact, while the share of U.S. light vehicle sales represented by imports from Asia peaked in 2009, the share of U.S. sales represented by vehicles produced in North America by Asian headquartered carmakers continued to rise through 2010, barely showing a decline in 2011. By the end of February 2012, that share had reached 30.4%, surpassing the previous peak from 2009 (29.9%). In other words, the share of U.S sales represented by North American production of foreign headquartered carmakers did not experience the same decline as that of vehicles imported from Asia.
Combining both of these developments, rising exports of new vehicles, up 50% since 2009, and declining imports of light vehicles, nearly flat since declining by a million units between 2007 and 2009, have resulted in the smallest U.S. trade deficit in light vehicles since 1998 (see figure 5).
Data on imports, sales, and production, are from Ward’s Auto Group, Auto Infobank, online database; data on exports are from the United States International Trade Commission website. Note that the definitions in both data sets don’t match perfectly. Ward’s data are based on individual vehicle models and their size classes. The trade data on vehicle imports are defined, at the most disaggregated level, by fourteen 10-digit codes and distinguish passenger vehicles from vehicles for the transport of goods, engine type, and engine capacity.(Return to text)
March 15, 2012
Export Effects of a European Slowdown on the Midwest
By Britton Lombardi and Bill Testa
As the U.S. economy has shown signs of recovery, attention has shifted to the European sovereign debt crisis and its impact on European economic activity, along with its spillover effects on the rest of the world. In the Midwest, where the manufacturing sector has been leading the ongoing economic recovery, concern has arisen that another recession in Europe could dampen economic activity here too.
What effect might a European recession have on international trade and the manufacturing sector in the Midwest? What do we know from past experience about the sensitivity of trade to changes in economic activity? A recent paper from Chicago Fed economist Meredith Crowley studied the causes of recent trade collapses, especially the Great Trade Collapse during 2008 and 2009. Crowley tested three factors that could potentially affect international trade: declining aggregate demand, financing difficulties, and rising trade barriers. In the past five of six U.S. recessions, the U.S. and the world suffered trade declines as economic activity fell in the two to four quarters before the trough of these recessions. Crowley notes that the elasticity of imports for the U.S. ranged from 1.5 to slightly more than 2 , which implies that imports respond more than proportionally to changes in income and demand (an elasticity of 1 would indicate a proportional response). Based on her own and others’ research, she concludes that a drop in demand is the most important factor in determining trade declines. Trade financing exerts a more modest affect on trade, while rising trade barriers have not been an issue recently.
We might expect similar declines in European demand for imported goods from the U.S. and elsewhere in response to a recession there. How would such developments be felt outside of Europe? To get a sense of how the Midwest, specifically the Seventh District, might be affected, we can ask how exposed the region currently is to trade with Europe—since any drop-off in trade is likely to be somewhat in proportion to current trading patterns. The Seventh District comprises all of Iowa and most of Illinois, Indiana, Michigan and Wisconsin. As seen below in Chart 1, exports to Europe grew for both the U.S. and the Seventh District during the 2000s, but still accounted for less than 2% of GDP for the U.S. and of gross state product (GSP) for the Seventh District in 2010. Therefore, exports to Europe account for a relatively small portion of the region’s economic activity. For 2010, the region’s exports to Europe were worth $31.5 billion, including agriculture and livestock products, oil, gas, minerals, and ores, as well as manufactured goods.
Comparing U.S. regions (Chart 2), the Great Lakes region (which includes Illinois, Indiana, Michigan, and Wisconsin, but adds Ohio and excludes Iowa) lies close to the middle in terms of export exposure to Europe. The Seventh District’s export-to-GSP percentage is almost identical to that of the Great Lakes region, at 1.88% versus 1.91%. The Mideast has the greatest exposure to Europe, with exports accounting for 2.65% of GSP, while the Plains have the least exposure, with only about 1.4% of GSP going to Europe.
What do we export to Europe? As shown in Chart 3, chemicals, machinery and transportation equipment, and computer and electronics account for about 70% of all Seventh District exports to Europe. How does the District’s overall export mix compare with other U.S. regions’?
The Seventh District’s exports to Europe tend to be more heavily concentrated in three product categories—chemicals, nonelectrical machinery, and transportation equipment. Of our top 10 categories, we overlap with the rest of the U.S. in six of them (Table 1, non-overlapping categories highlighted in red).
More importantly, our top trade categories are in sectors (industrial goods, such as machinery and equipment, and industrial chemicals) whose sales tend to rise and fall disproportionately with swings in overall business activity. A recent National Bureau of Economic Research working paper noted the procyclical nature of traded goods, specifically durable manufactured goods, and again found that the majority of the decline in international trade was due to a decline in demand for manufactured goods. The authors also find that the decline in the trade of durable goods versus nondurable goods (soft goods like food and clothing) accounted for a larger portion of the decline in total manufactured goods demand in both the recent Great Trade Collapse and the 2001 recession. Therefore, the Seventh District’s reliance on nonelectrical machinery as a top export (21.1% of total exports) could make our region somewhat more susceptible to the negative effects of a European recession than the rest of the U.S., where nonelectrical machinery accounts for a smaller proportion of exports (7.4%).
So, although the Seventh District economy’s exposure to Europe through the export channel is somewhat limited, a slowdown in Europe may be expected to hurt our region’s exports more than those of other regions that export less sensitive products. Exports and imports have historically been sensitive to overall economic conditions. In some cases, such as the 2008–09 global recession, trade declines can be severe.
Export data is pulled from the International Trade Administration’s TradeStats Express. Of total Seventh District world manufacturing exports, Seventh District sends about 48% to North America (Mexico and Canada), 19% to Europe, 17% to Asia and 7% to South America. For information on the District’s major trading partners by individual nation, see this blog. (Return to text)
March 1, 2012
Seventh District Update
by Norman Wang and Scott Brave
A summary of economic conditions in the Seventh District from the latest release of the Beige Book and from other indicators of regional business activity:
• Overall conditions: Economic activity in the Seventh District continued to expand at a moderate pace in January and early February.
• Consumer spending: Growth in consumer spending slowed in January and early February. Contacts indicated that activity was boosted by clearance sales and also some isolated improvement in the luxury segment. Auto sales were up in January, but down slightly in early February.
• Business Spending: Business spending increased in January and early February and inventories were indicated to be at comfortable levels. Labor market conditions improved, although hiring remained selective.
• Construction and Real Estate: Construction activity was up slightly in January and early February as multi-family construction remained an area of strength and nonresidential construction continued to trend up moderately. Commercial real estate conditions improved with vacancy rates edging lower.
• Manufacturing: Manufacturing production increased further in January and early February. Demand for heavy equipment remained strong and the auto industry also continued to be a source of strength. Manufacturers of specialty metals reported solid order books.
• Banking and finance: Credit conditions were slightly improved from the previous reporting period. Financial market volatility declined and risk premia moved lower across a number of asset classes. Banking contacts indicated that loan growth continued at a moderate pace with demand from larger businesses being stronger than that from small to mid-sized companies.
• Prices and Costs: Cost pressures were largely unchanged in January and early February, but the volatility of commodity prices remained a concern for many contacts. Wage pressures continued to be moderate.
• Agriculture: Corn, soybean, wheat, hog, and cattle prices rose during January and early February. Input costs for agriculture continued to increase, led by sharply higher rental rates for cropland.
The Midwest Economy Index (MEI) increased to +0.09 in December from –0.13 in November, rising above its historical trend for the first time in five months. Midwest growth also outperformed its historical deviation with respect to national growth, as the relative MEI increased to +0.31 in December from –0.06 in November.
The Chicago Fed Midwest Manufacturing Index (CFMMI) increased 1.7% in December, to a seasonally adjusted level of 87.4 (2007 = 100). December’s growth was broad-based; all four major industry sectors of the index indicated growth in production activity. Revised data show the index was unchanged in November. The Federal Reserve Board’s industrial production index for manufacturing (IPMFG) increased 1.5% in December. Regional output in December rose 8.4% from a year earlier, and national output increased 4.6%.