Robert Gibbs
Economic Research Service , U.S. Department of Agriculture

I'm going to do a Power Point Presentation (Education and the Workforce: Delmarva in the Rural-Urban Context --PowerPoint file), but I did want to make a couple of points about "what is rural." It's an important issue; and it's not been settled. I think it's good to know that in some ways you are comfortable with where you fit on the rural-urban spectrum. That's not true of every place. So I wanted to bring up the point that, in terms of the new census classification system, as of 2000, Salisbury is no longer technically a "non-metro" place, that Wicomico County is no longer a nonmetro area. And also to alert you, because it may have implications for the way people perceive the region, and maybe even the way you perceive yourselves, there's a new intermediate classification, called "micropolitan" area, introduced by the Office of Management and Budget, which essentially is defined in the same way that “metro” area is, except that instead of requiring a city to have 50,000 people, this center only has to have 10,000 people. Cambridge, in Dorchester, Sussex, and Worcester, called Ocean Pines I think, also are included. So that could have an effect on the way people view this part of the Eastern Shore. People, especially those with budget authority, should see what's happening here and how the area is changing.

Most of what I'm going to show you now is a rehashing of what Dan and Memo presented this morning, maybe in a different guise. I put this together because it helped me to think about the issues that you're dealing with, and what I wanted to talk about today. I learned a lot about Delmarva and rural education and economic development by putting this together.

I call it "Delmarva in the Rural-Urban Context," because I'd like to show you comparisons with other rural and urban areas. I broke out the industry mix of the six counties that you're concerned with and compared it with what's going on overall in the rural and urban economy (we call them "non-metro" and "metro.").

So (slide 1) Delmarva doesn't fit the typical pattern of either. It depends on the industry you're talking about. In agriculture, for instance, it's a little less concentrated in employment. Delmarva falls between a typical rural and a typical urban profile. Results might be different if you look at income, or in other ways. If you look at manufacturing, for example, Delmarva is much more like a typical urban profile, about 13% employment, which is much less than a typical rural area, where manufacturing is much more important for the last several decades. Is this a transition from a much higher level of employment? Is it moving toward an urban profile because skill levels are rising and manufacturing productivity is going faster here, and so employment tends to shrink in the face of that? Or are there other factors at work?

In the retail sector, the arts/entertainment/recreation sector, Delmarva comes out above both the rural and urban profile. That is not surprising, given that you have a center like Salisbury, and with the retail sector that's very vibrant along the coast in Ocean City, Rehoboth, and places like that. And the high employment rate isn't surprising, given your concentration in tourism. Professional services does fall off the urban profile; employment is only half of what you get in a typical metro area. So, all in all, you have a real mix here.

What I did then was look at educational attainment (slide 2). Delmarva fits more with a typical rural county. The share of adults 25 years or older, with at least 4 years of college or an advanced degree, on Delmarva is a couple points higher, several percentage points above the rural (with 15.5% and 17.5%); but it's below the metro educational attainment of 26.5%. High school dropouts matches the rural profile almost perfectly. If you calculate the educational level needed in each of the industries we talked about, and apply those average education requirements to Delmarva's industry mix, you'd expect Delmarva to have a higher educational level than it does. I'm not convinced it is just due to the mix of specific industries you have here because there are other things that play in determining what the educational level of a population is, like the age structure. With a higher share of older residents, you'll have lower attainment, because each cohort has been rising in educational attainment. It's also true, and important for Delmarva, that racial and ethnic composition plays a significant role here. Educational attainment varies quite a bit across the different counties. It's lower in Accomack, Dorchester, and Somerset counties. It's higher in Wicomico and Worcester.

The next thing I did, just for fun, but interesting, I went out to all counties in the United States and tried to find a twin for each of the six Delmarva counties. I did it by looking at each county's industry mix and then finding the county that matched it with the closest mix. I used the Index of Dissimilarities, which compares all 3,000 counties. I thought they'd be nearby; but they're not, at all.

Accomack County 's twin is Grady County, Georgia, which is in south Georgia on the Florida border. Dorchester County 's twin is Canyon County, Idaho, which is just west of Boise. Somerset County 's twin is Pittsburgh, Oklahoma, in the southeast area of Oklahoma. Sussex County 's twin is Benze, Michigan, south of Grand Travers Bay, right on Lake Michigan. There's quite a bit of recreation and tourism there, too. Wicomico County 's twin is Cape Girardeau, Missouri, south of St. Louis on the Mississippi River. Cape Girardeau may be the county seat, but its population profile matches Salisbury 's, to a degree, and also is a regionally important center. Worcester County 's twin is Harrison, Mississippi, on the Gulf coast near Biloxi and Gulf Port, where there is lots of tourism and recreation.

Since they do match on industry mix, I looked at the educational attainment to see how they compare. The share of adults who have not completed high school in 2000: the pattern is Delmarva has a higher share of high school dropouts. You have a similar set of jobs, but a slightly different educational mix. And it's quite different in Somerset and Sussex. Worcester is the only one to break out of this pattern. It is slightly lower than its Mississippi twin. That's not surprising. For those with at least a four-year degree, Delmarva has a lower share of educational attainment, except Accomack and Worcester. So, again, the two deep south counties didn't do quite as well. There definitely are regional and, I would argue, a demographic component to all of this.

There are a couple of topics I'd guess you'd want to keep in mind as you develop an education development strategy. First is your metropolitan proximity. Delmarva is unique, for rural counties.

It's so close to large, major metro areas, such as Philadelphia, DC, Baltimore, and Hampton Roads. And given the large occupational industry mix difference between those places, Philadelphia and DC aren't typical metro areas. They have a very high rate of professional and managerial employment. So it's not hard to see why you'd have a hard time overcoming a long-standing brain drain. This is more of a challenge than in other rural areas, which may have equivalent opportunities in their local area, and whose young people don't have the same set of nearby opportunities to pull them away. So it's easier here just to leave for a lucrative alternative.

The flip side of this, and increasingly important in rural development, is that you do have these places with huge population areas nearby, with folks -- managers and professional -- looking for places with high levels of natural amenities and who probably have experience with this area through vacations and visiting here. So you have a real strength here to have proximity to a population with financial resources to consider Delmarva as a place to retire to or set up a second career, as telecommuting becomes more real and commonplace. But high quality schools are a key location factor for this group. Schools are seen not just as a place where you provide kids who are here with tools to go out and do well with a career; it's become a selling point for rural areas, and there are many who are weak. It's a lot worse in many places. This is a great selling point, especially to the group of folks who don't live so far away and who are looking around for a place to retire to or start a second career.

This is just to show you: there's a huge disparity between local earnings and those in metro areas nearby. "Earnings/job in Manufacturing, in a Retail Industry, in Year 2000." I picked Baltimore County and Chester County, Pennsylvania for comparison. The first set of bars represents manufacturing earnings, which generally shows more variation across states. The second bar represents retail earnings, which is much more similar, wherever you go. Even for retail, we see differences between Wicomico ($15,300) and Chester County ($26,100). These figures come from the Department of Commerce. Living costs are higher than in Wicomico; but this is an area where there's still lots of land that is fairly moderate and reasonable in house values. The indication to me is that there's a real welfare difference, as an economist would say. So for a young person who's non-college-bound, and is thinking about options in the short term, this is the best place to get a return on the education they have. There are some options that aren't so far away, and this is just one place. Suburban counties similar to this are all over the East Coast. And of course, manufacturing is even more varied, and there's more disparity – Wicomico County ($37,000) and Chester County ($78,000). One of the qualifications I'd put on manufacturing is that while disparity is there, I wouldn't advise a student to get a job in Chester because that sector hasn't been doing too well for a whole range of things. You see the high earnings coming from unionized labor; but these kinds of shops are quickly disappearing. So it's not realistic in all cases. But this is indicative of the calculus that they're making and why it's a challenge to Delmarva to retain these young people.

The other thing we talked about this morning was the racial and ethnic inequalities. This morning we talked about the Hispanic, Korean, and East European burgeoning populations. But the traditional diversity on Delmarva has been between whites and African-Americans. And Delmarva still has a large share of African-Americans. Hispanic growth has been tremendous in the last 10-20 years, but it's a small share, overall. So it's a future issue, but it's not happening right now. Somerset is 41%, Accomack is 31.8%, African-American. And I think the same figure that Memo had us write down is the same. I think this is an issue you really can't get around. You might be saying, “You're showing us educational attainment of adults and their educational choices have already been made, so, what difference does it make?” The last generation is a very important predictor of what's going to happen in the next generation. It's because parents' education attitudes toward school and their income are critical predictors of a student's future attainment. And when you have a population that has a very different educational attainment level, that's a challenge. In 2000, 7% of African-Americans, 20% of non-Hispanic whites, and 11.6% of Hispanics 25 or older have a college degree. I'm not persuaded by that statistic for Hispanics. Maybe some of you could tell me a reason for that. I think it's a small sample size, which may be driving it. For those without a high school diploma: 39% of African-Americans, 58% of Hispanics, and 19% of non-Hispanic whites are included. To me, this is a challenge that needs to be addressed at some level. And I credit the discussion today with saying that diversity issues are going to be a major issue in the community -- not just the growing population, but the population that you already have in place.

I'm a firm believer that you shouldn't discount manufacturing. Some say it's going the way of agriculture; the share of employment in it is declining. But over 20% of employment in rural areas, overall, still comes from manufacturing. It's still a very key anchor sector in many rural economies. I think that's true for much of Delmarva as well, even though it only makes up 13% of the employment here overall. It's very important as a base. More importantly, manufacturing job growth in the 1990's, unlike earlier periods, really is related more closely to higher educational attainment levels.

I have a slide I can't take credit for, by David McGrannahan at ERS, who came up with the slide. If you look in the 1980's, this is the change in manufacturing jobs. We took all the nonmetro counties and ranked them according to educational attainment level. The first bar represents counties in the bottom half of the education mix. In the 80's, we had 6.5% growth in manufacturing jobs, where the higher education counties had only 3% growth. In the 90's, that reversed in lower education areas. There was a 3% drop, but an almost 9% increase in higher education areas. So, an educated labor force is much more important for a location. Manufacturers say a skilled labor pool is important; and very often they were having trouble finding it. They can get them trained, but they can't keep them, an extension director in a small town in Indiana told me.