A couple of weeks ago I posted an appeal by reader Sam MacDonald, making a case for Benedict Option folks to relocate to his part of the world, Elk County, Pa.  Sam’s piece caught the attention of Salim Furth, an economist at the Heritage Foundation, who sent me this essay:


Rod Dreher posted a long dispatch from Sam MacDonald, president of a busy Catholic school system in Elk County, Pennsylvania. MacDonald reports that a thriving, social-spiritual community of Catholics has carved out a niche in the beautiful hills of northern Pennsylvania. The quote from McDonald’s letter that I tweeted emphasized the economic opportunity:

Last week there was an ad in the paper for an entry level worker at one of the local plants. GED required. It started at $15.60 an hour. A greenhorn at the papermill can expect to earn upwards of $50,000 first year.

Cost of living? If you walked into Ridgway tomorrow and offered my mom $40,000 for the house she raised four kids in, she’d probably take it.

With mounting evidence that social and economic health go hand in hand, it would not be surprising if an enclave of good parish schools and well-connected community organizations fostered high-productivity jobs.

But is Elk County really special? Would data on social and economic outcomes support the anecdotes of a self-admitted local booster?

Sadly, no.

To be sure, Elk County is really Catholic. Really, really Catholic. Seventy percent Catholic. Elk also has the second-fewest evangelical Protestants in the state. Elk is unique on another dimension: it is the most Caucasian county in the state, at 97.8 percent white, Mayor Guillermo Udarbe notwithstanding.

Other than the cassocks, Elk County looks a lot like any other rural Pennsylvania county.

Average income in Elk County is normal for the region. The number of social associations per capita in Elk County is just a bit above average. Unemployment, loosely estimated at 6.7 percent, is substantially higher than in Pittsburgh or metropolitan Philadelphia, but typical of rural Pennsylvania. Elk County’s population is declining – which might explain why factories are having difficulty finding labor.

Demographically, I compared Elk County to 43 other Pennsylvania counties that are 90 percent or more white (which effectively drops the cities and suburbs). Elk has a typical rural demographic structure, with 20 percent of the population under age 18 and 20 percent aged 65 and up.

Family structure is mostly normal. Almost 80 percent of family households are “husband and wife families”, as is the case elsewhere. The one oddity, perhaps, is the lack of large families. There are just 86 households in Elk County with at least seven members. That’s 0.6 percent of all Elk County households. By contrast, wealthy, urban Arlington, Virginia has 0.9 percent large households and suburban Loudoun County, Virginia has 2.1 percent large households. As a result, average family size in Elk County is among the lowest in rural Pennsylvania, although not by much.

Elk County denizens succumb to temptation a little less than average. According to County Health Rankings, Elk’s smoking and excessive drinking rates are typical of Pennsylvania. Elk is a moderate shade of orange on the dramatic national map of drug overdose rates. With about 15 overdoses a year per 100,000 residents, Elk is seriously affected by the national heroin epidemic, but is not ground zero. Elk has one of the state’s lowest rates of chlamydia infection, which suggests low sexual promiscuity, but the teen birth rate is normal. Elk’s brightest spot is its violent crime rate, which is the third-lowest among the comparison counties.

There’s scant evidence that Elk County is a uniquely cohesive community with particularly strong institutions. The dearth of large families suggests that the hardcore Catholic subculture is very small indeed. The simplest explanation that squares the data with MacDonald’s observations is that the dégringolade of rural America is rather overblown. The moral of the story is not that Elk County is a mess – it’s not – but that most other counties are in pretty good shape, too.

Elk County does, however, have one statistical advantage that might make suburbanites pull up and move there: the shortest commutes of any county in Pennsylvania.

I offered Sam (a former writer for Reason, by the way) the chance to respond. He writes:

I would like to thank Rod for giving me an opportunity to respond to what I think is some misdirected analysis here. I would also like to thank Dr. Salim Furth for engaging with my piece in the first place, and in most cases making a stronger case for my position than I ever could.

I stand by my original claim that Elk County is a location uniquely well-suited to the Benedict Option. I based it on a few things, including the fact that the area is deeply Catholic, that it’s safe, that it’s cheap, and that a middle class existence is well within reach even for people who did not go to college and instead work industrial jobs.

Furth concedes a great deal of my claim. In his words, “Elk’s brightest spot is its violent crime rate, which is the third-lowest among the comparison counties.” He also says, “To be sure, Elk County is really Catholic. Really, really Catholic.”

He then makes some odd moves which are hard to analyze because he does not provide links to his data. On the discussion of family structure he says that in Elk County, “Almost 80 percent of family households are ‘husband and wife families’, as is the case elsewhere.” Data here suggest that the statewide percentage is about 75 percent. Elk County is higher, but not by a large amount, which would seem to be based on the fact that there is not an awful lot of room to be dramatically higher.

Furth then launches into a confusing discussion of household size, which quickly morphs into a discussion of family size, which are two different things. Again, it’s hard to analyze because he doesn’t provide any links. He then makes the strange move of comparing household size in Elk County to what he admits is a large, wealthy county like Arlington in Northern Virginia. As this article explains, Arlington is a uniquely bad point of comparison. There has been a huge population boom there as Washington, DC, continues to sprawl. At the same time, much of the affordable housing has been removed, just as a huge wave of foreign immigrants has been imported to fill the booming, low-wage retail sector. So faced with relatively low wages and a staggering rate of income equality, the poor people do the sensible thing and cram a lot of people into small spaces.

Why don’t struggling people in Elk County cram tons of people into housing? Well, because they don’t need to. I searched Zillow for houses in the $240,000 range in Arlington. Generally speaking you are looking at one bedroom condos of less than 1,000 sf. Here is a house that recently was on the market in Ridgway for $240,000. It never sold at that price and was taken off the market. It says four bedrooms but it’s really more like six. It’s $3,500 square feet. So again, I am unclear what household size in dramatically different places is supposed to tell us about Catholicity in Elk County.

Of course, it might be convenient if a major, county-by-county study had been done to address the question of religion in America so we wouldn’t have to follow Furth’s idea of judging Catholicity by using household size as a proxy. I’m not an economist at the Heritage Foundation, but I do know what Google is and someone did do such a study. The Washington Post crunched all that data and built interactive maps. The story was picked up by none other than Rod Dreher himself. The one about religious participation struck him so profoundly that he dedicated a whole post to it.

Rod’s post bemoans the lack of religious intensity. But scroll down. A commenter named Sam M (me) says look at the map. There are only a small handful of counties with intense religious participation east of, say, Indiana. You can see it quite easily. A blood red county in a sea of beige in western PA. That’s Elk County. The links to the interactive version are broken at the Washington Post’s site, but I preserved the data about Elk County in the comment. Out of 3,143 counties in the US, Elk County ranked 101 in terms of religious participation–meaning the number of people who actually practice the faith they profess to follow. That puts us in the top 3.3 percent nationally, and in a league with intensely Mormon and evangelical regions in Texas and Utah.

So there is no need to guess. The data say Elk County has a crazily high number of Catholics, and a tremendously high percentage of them actually practice their faith. I would also invite Furth to ground truth the data a little. One of the kids on my boy’s peewee football team is from a family of ten children. We will soon welcome our eighth. Within a five minute walk of my house are families with six, seven, ten, eight and eight. And many more with three to five. Maybe it’s like that in Arlington. I don’t know. I don’t live there. I doubt I could afford a house big enough if I did.

I get a little more nervous when we turn to the economy, because of course Dr. Furth is an economist. But it’s here where I find his analysis most perplexing. I claim that Elk County has a thriving middle class, which is definitely reachable for people who hold one of the many industrial jobs on offer. And I claim that people who hold these jobs are respected, and that their children will have plenty of opportunity to become whatever they want to be in life. Which is really a claim about class and mobility. Furth makes no effort to engage with that claim, except to point to a few basic metrics such as the unemployment rate and average income.

He says, “Unemployment, loosely estimated at 6.7 percent, is substantially higher than in Pittsburgh or metropolitan Philadelphia, but typical of rural Pennsylvania. Elk County’s population is declining – which might explain why factories are having difficulty finding labor.” Furth’s analysis might have benefited from picking up the phone and actually asking someone. The decline in population is almost ancillary to the employment crunch. Population is down, but the factories could have weathered those demographics because individual workers are dramatically more productive than they were 20 or 30 years ago. These are world-class people. There is a good chance that the Michael Jordan of machinists lives here. The Albert Einstein of tool and die makers, too. That’s why they get paid a lot.

Instead, culture is a large part of the problem. I started high school in 1987. Know what happened in 1986? The Homestead Works shut down in Pittsburgh, just a few hours away. Even three years before that we had movies like “All the Right Moves,” featuring a young Tom Cruise desperate to get out of his crappy western Pennsylvania mill town. We all knew, because our parents told us, that if we could get out, we should. A life of riches and independence awaited all those who went to college. So guess what? We did. We left. The problem now is that the guy who was a 40 year old tool and die maker in 1987 is now getting ready to retire and there’s nobody to replace him.

Moreover, there is the issue of work ethic. We are not nearly as bad as the community highlighted in Hillbilly Elegy, thank God. But finding people who can pass a drug test and show up beyond the first paycheck is becoming a challenge. Why? Because many of the people who would have been great at these jobs were steered to college and told that working in a plant amounted to failure. That was a terrible mistake. Which is why local industrialists are making long term plans to steer our best and brightest back into industry, but also find people to relocate.

Again, I am not an economist on par with Furth, so it would be really useful if some highly respected economists would do a county-by-county study that highlighted all the issues related to social mobility that I have been talking about.

Oh look. Someone did. That’s a huge New York Times article discussing a major study the paper identifies as, “the most detailed portrait yet of income mobility in the United States.” It was done by some guys at Harvard and UC Berkeley, one of whom is Raj Chetty, recently named “the country’s best academic economist under the age of 40.” So no. I am not an economist. But he is.

The study indicates that Pennsylvania is actually one of the top states in terms of income mobility, with urban areas such as Pittsburgh ranking very high. But the study wasn’t limited to urban areas. The Keystone Research Center helpfully jumped into the study to see how all regions of the state ranked on three measures of mobility, and on the percentage of families achieving middle class status. Take it away, Keystone!

St. Marys stands out among all the Pennsylvania regions on all three of our upward mobility measures. The chance of a St. Marys child in the bottom fifth rising to the top fifth is 12.9%. The average income percentile of St. Marys children who grow up at the 10th percentile is 44. And St. Marys children whose parents’ income were in the bottom half of the distribution have an average income percentile as adults of 48.4 — very nearly the U.S. median. Appendix Table A1also shows that St. Marys has a stunningly large middle class — 70% of children in St. Marys grew up in households with incomes between the U.S. 25th percentile and the U.S. 75th percentile.

As a matter of fact, the St. Marys region led the state or tied for the lead in every category measured.

So. If you live in St. Marys there’s an astonishingly high chance that you will be in the middle class. On the off chance you are poor, you will have a much better chance of rising to national median than most places in the country. Notwithstanding Furth’s reliance on blunt, context-free analysis of unemployment data, I stand by my claims.

Of course, if it’s income he wants to talk about, we will. Furth seems confident that income in Elk County is on par with other rural counties. Let’s go to census data and measure Elk against neighboring counties, and also see what that income will get you in terms of housing:


Median Household Income Median Selected Monthly Owner Cost, with a Mortgage Median Gross Rent
Elk $46,576 $962 $389
Cameron $41,157 $948 $374
McKean $42,913 $910 $368
Clearfield $41,510 $1,017 $407
Forest $36,037 $860 $322
Jefferson $42,295 $936 $364
AVERAGE $41,748 $939 $371


I am using median household income. That’s what the census most readily reports. Furth did average income. Maybe he has a rationale for that, but he didn’t explain it or provide links to the data. So going on what I have, we see that the cost of home ownership is slightly higher than average in Elk County, which would cost you $264 more per year. Rental, you’d spend an extra $216 a year, on average. Of course offsetting that is the median income, which is dramatically higher than average by $4,828. That’s 11.5 percent higher than the regional average. I doubt $4,828 would go very far in Arlington, Virginia, where the median value of owner occupied housing is $594,800, but it goes a very long way in northcentral Pennsylvania.

Now, regarding Furth’s analysis of social measures, he admits that measures of promiscuity look pretty good. And he then goes into some discussion of drinking and drug use. But before we even get into that, I hope it’s not too indelicate for me, as a part of the tribe, to ask if Furth, uh… knows what a Catholic is? I mean, there are jokes about us, right? We make them ourselves. Nobody claims that we are Baptists or Mormons. Elk County has the third highest number of bars per capita off all the counties in Pennsylvania, and we recently were declared the binge-drinking capital of the state. There is clearly some pathology in that fact, and the social service directors will caution you not to view that statistic with a smirk. Of course many locals do. The standard for binge-drinking is five drinks or more in a sitting for men. It’s not uncommon to see some of the best, most Catholic people in the world have that many around a horseshoe pit. So how much of that is pathology and how much is German Catholicity? Hard to say.

But most interesting to me is Furth’s claim that, “There’s scant evidence that Elk County is a uniquely cohesive community with particularly strong institutions.” One measure of this, which he again fails to link for further analysis, is that, “The number of social associations per capita in Elk County is just a bit above average.” I really do wish he had been more forthcoming because I honestly don’t even know what “social associations per capita” means. Neither does Google. A search of that exact phrase turns up one hit, from a letter to the editor published in Luzerne County. Maybe it’s important. I’d be willing to think it might be. I await further guidance.

But more important on this front is the phrase that does all the heavy lifting in Furth’s piece: “Other than the cassocks, Elk County looks a lot like any other rural Pennsylvania county.”

Wait. Other than the cassocks? The cassocks are the whole point, when you are discussing whether a given community would be a good Benedict Option for Catholics. It’s like saying, yeah, other than that whole pilgrimage thing, Mecca isn’t such a great place for Muslims to visit. Or gee, what’s there for a Catholic to see in Rome, other than the Vatican? Not much in that art museum, apart from the art. That pizza place has the world’s best pizza at extremely reasonable prices. Other than that, it’s terrible.

While Kurth is busy looking into mysterious “social associations per capita,” I am saying, hey fella, uh… we have a Catholic school system that draws close to 20 percent of the local school population. The local parishes chip in upwards of a million dollars per year to ensure that the school is accessible to folks in our (astonishingly large!) middle class. We have a huge number of Catholics, and an incredibly large portion of them are active participants in their churches.

Then the economics. The rock bottom housing prices. The median income which is well above regional average. Which combined with the low housing process mean that your money goes a long way. And the fact that in a state that ranks near the top of the national heap in social mobility, St. Marys ranks highest in all measures put forward for analysis.

So yes suppose. Dr. Furth is correct that there is very little data to support the claim that Elk County is different. Except, you know, all the data that do.

And if you really doubt that there are industrial jobs, look and see for yourself. Dr. Furth has a research position. Subscribe to the St. Marys Daily Press. I just picked up a week’s worth of papers off my desk at random and here is list of the jobs offered:

Product Engineer. Production supervisor. Experienced die setters (offering $1000 sign on bonus). Maintenance technician ($1,000 sign on bonus). Entry level operator positions (plural in the original). Manufacturing engineer. Electrical engineer. Diesetters. Secondary machine attendants. Maintenance mechanics and electricians. Quality engineers. Manufacturing engineers. Manufacturing planner and customer service rep.

Actually, that’s not a week’s worth. It’s one single day in late September.

This past Monday? Quality inspector. Immediate openings: Entry level labor. Production planner. Powdered metal sales and marketing manager. CNC set up. Experienced mold set ups and operators.

All this ignores the ads for teachers, nurses, speech pathologists, block layers, PT floral designer, Mechanic. Etc.

By all means, if Dr. Furth believes this is an illusion based on homerism and that a kid with a high school diploma and a few hundred dollars in his pocket—or a BenOp Christian looking to relocate–is just as likely to find a middle class life in Arlington, VA, as he is in Elk County, he should advise people according to his analysis. Me? I would suggest such a person move here instead.

About 30 people have already contacted me about doing just that.

I’m working on it!


This is a great discussion. Thanks to both men for it. If Dr. Furth wants to respond in kind, I’ll publish it here. Watch this space.