This chart explains why coastal elites and middle America see the economy differently

By most measures, the American economy is doing pretty well. Unemployment and inflation rates are low, and the economy has been generating jobs at a steady clip for more than five years.

Yet the current recovery doesn’t feel right to many Americans. A lot of people feel that media and political elites have become out of touch with the economic struggles they face.

A new report from the Economic Innovation Group sheds a lot of light on what’s going on. One chart in particular explains why people in big cities tend to have a very different view of the economy than people in more sparsely populated parts of the country:

The chart shows the growth of establishments — that is, business locations like restaurants, car repair shops, or factories — during the first five years of the past three recoveries. It shows that the current recovery has been slower than the 2002 or 1992 recoveries overall, but the more interesting fact about the chart is how this growth is distributed across the country.

In the past, smaller counties tended to grow faster than larger counties. This made a certain amount of sense — large counties like Los Angeles or Dallas were already expensive and crowded places to live, so it was easier for economic growth to happen in smaller towns or outlying suburbs.

But in the latest recovery, the pattern has reversed. The largest counties saw the fastest growth in establishments, with Los Angeles County, Miami-Dade County, and Kings County (Brooklyn) leading the way. Meanwhile, the least populous counties haven’t enjoyed a recovery at all: Businesses continued to close stores and factories, on average, in counties with fewer than 100,000 people in them.

You can see a similar pattern for job growth:

Small counties gained jobs at a faster rate than average during the 1990s and about as fast as the national average between 2002 and 2006. But in the current recovery, small counties have been left far behind, gaining jobs at about half the rate of big cities.

It’s hard to say exactly why this is happening. In part, it reflects a long-term trend toward urbanization. Another important factor is the shift toward high-tech industries like information technology and biotechnology. These industries thrive in big cities where they can draw on deep pools of talent.

And this suggests that the housing shortages in cities like San Francisco, New York, and Los Angeles are likely a major impediment to nationwide economic growth. In previous decades, booming economies in Sun Belt cities drove rapid population growth in those cities, contributing to the rapid growth of the economy as a whole. In contrast, the San Francisco Bay Area, the nation’s most dynamic metropolitan area, grew an anemic 1.4 percent per year between 2010 and 2015.

The reason for this is that excessive housing regulations make it difficult to significantly expand the housing stock in San Francisco, New York, Boston, or other major cities. There are fast-growing companies in these cities that would like to hire more people, and there are people outside these cities that would like to take the jobs. The problem is that there’s nowhere for them to live.

And faster housing growth wouldn’t just mean more jobs at high-tech companies; it would also have spillover benefits up and down the income scale. If Facebook were able to hire more engineers, those engineers would want demand the service of more dry cleaners, chefs, doctors, and teachers. That would allow a lot more people who are currently struggling to find work in smaller towns and rural areas find a new, more prosperous career in a big city.