A Population History of Denver
Or: Acts of God Are Usually Benevolent
I have argued before that relocation due to natural disasters usually has benevolent effects for the relocated people, most recently mentioned here, and the most recent academic confirmation of this frequent finding is a fun paper about volcanoes wiping out Icelandic fishing villages.
But what about the impact of natural disasters on destroyed places? Right now, Louisiana and Mississippi are enduring historic floods that have destroyed hundreds of thousands of houses. Quick note here: anyone who believes that there are no sudden, violent shocks to housing supply should wake up and smell the swamp-water. When you consider the role of disaster/repair cycles, housing supply can be very volatile. Look at New Orleans’ housing stock around Katrina for an example.
This question is going to be a big deal for today’s topic: Denver!
I’m fond of Colorado due to childhood summers spent first around Estes Park, and later in Buena Vista, Colorado, where I have family. But I know very little about Denver. I realized this when a response to another post went into a tangent on Denver and I realized, huh, I dunno much about it. This is odd, since Colorado is actually kind of a big gainer state from domestic migration, yet I really have no idea why specifically, nor where.
So with that, let’s explore Denver.
Visualizing Denver’s Population History
The above chart shows annual population for Denver City from 1860 to 2015, as well as the sum population for all the other counties in the current metro area from 1900–2014. I want to digress here on a method issue: backcasting current MSA boundaries is a neat trick that creates an appearance of different growth than may have occurred. The Denver area has seen many expansions of its MSA boundaries, but has also seen several counties spun off into their own MSAs, such as Boulder and Greeley. To the extent that lost counties may have different growth trajectories, the growth or decline in what was viewed as “Denver” in 1980 may be different from the growth or decline I present here.
Furthermore, the newest county in America, Broomfield County, is part of the Denver MSA, which serves as a reminder that county definitions change over time. Finally, I do the surrounding counties back to 1900 because, before 1900, we have some seriously wonky counties, and back in the 1860s we have counties bigger than the future state of Colorado. Plus, counties in easy driving range of Denver today were not in easy travel of Denver in, say, 1900, meaning that backcasting today’s metro area exaggerates the extent to which historical populations in the same counties were actually connected to Denver.
All that is to say: I like technical details and like to show you the man behind the curtain so to speak, but, in the end, I have no corrective. There’s no great way to track “population of the area seen as the Denver area in a given year” because that involves adding or dropping whole counties at arbitrary times, creating a broken and inconsistent time series. No bueno. As such, back-casting the current MSA is as good as we can do for now, but just be aware, it’s a flawed method for exploring historic patterns.
Luckily, Denver itself is a discrete jurisdiction, so we actually can meaningfully track the population of the Denver municipality, and late the unified Denver city/county.
Denver’s Early Years
Okay, so, what do we make of Denver’s population?
We clearly have early growth in the 1800s. That’s due to Denver’s status as a mail distribution hub, a state capital, a service town for Colorado’s booming mines, a market town for Colorado’s booming agriculture, and one of the few cities in the area with a commercial and passenger rail spur. Fun fact, the rail spur was created thanks to locals pooling cash to make a line from the Transcontinental, which ran through Cheyenne. They worried that without a spur line, Cheyenne would supplant them as the major mining headquarters, agricultural service, and mail distribution center. As you’ll see, Denver’s urban commitment to ensuring its logistical centrality will be a recurring theme.
But then there’s a sharp decline around 1893–1897. This is due to the Panic of 1893, and a sharp drop in commodity prices. Denver’s economy dropped through the floor. The population fell by nearly 12,000 people. But this disaster did not break Denver! Instead, the local community banded together and founded a series of community organizations and charitable societies, including the original precursor to United Way. Impressive.
Oh, and they also founded race-nationalist groups intended to ensure that native-born Americans got jobs, but Catholics and foreigners didn’t. Several thousand immigrants and Catholics were laid off during the depression and replaced with native-born Protestants. This episode of racism was not isolated, and by the 1920s there would be KKK members in high office throughout Denver and Colorado.
The growth you see after the boom is also artificial: Denver rapidly expanded its municipal boundaries during this period, growing in technical population despite not actually seeing any specific community grow.
But, by 1900, real growth had begun again. By this time, Denver was a large city, known for its healthy air and good hospitals and sanitoriums, its exciting night life and amusements for miners and cowboys come into town with their earnings, its opulent hotels and mansions for the wealthy mine-owners, and its faster access west than most nearby locations.
Through WWI the city continued to see growth, but it also developed a criminal reputation: mobsters, gamblers, and the seedy underbelly of the “wild west” came to dominate urban politics, even as a parallel network of charities, hospitals, religious organizations, and benevolence societies tried to promote a different vision of the city.
What both groups had in common was an ability to mobilize Denver residents for key projects: like the construction of the Moffat Tunnel. This tunnel, completed in 1928, dramatically shortened the rail distance from Denver to the coast and, despite being quite a long ways away from Denver, was, again, paid for by Denver’s residents, government, and businesses. Once again, the city worked hard to secure its logistical centrality.
And it’s a good thing they did. By the mid 1920s, Denver’s population growth had begun to slow, and the first inklings of suburbanization had begun out beyond city limits, with railways and cars starting to let some people live further out from the core. Likewise, cars began to create a new problem to add to Denver’s vice, crime, and corruption: smog.
The City Needs a Bath
When I wrote about Atlantic City, I explained how a culture of corruption and crime took over the city and destroyed its most valuable economic assets. The same thing could have happened in Denver. However, it was saved by a combination of factors.
First, prohibition was better enforced in Denver than in Atlantic City, though still quite imperfectly. Second, Colorado had a wide variety of legal and illegal gambling establishments, so the vice of Denver didn’t have nearly the monopoly that Atlantic City did. Third, Denver didn’t have major cities like New York nearby to make the vice economy large enough to completely sustain the city’s economy, thus other economic sectors maintained real sway. Fourth, unlike Atlantic City, Denver got major Federal investments during the 1930s and 1940s. A series of New Deal projects had some positive impact, but the permanent location of a slew of Federal agencies in and around Denver in the 1940s had a more lasting impact. Likewise, as Denver was far from the coasts, it was seen as an ideal secure location for sensitive military facilities, as well as munitions factories.
Thus, in the leadup to WWII and during its aftermath, Denver benefited from a huge number of Federal employees: as much as 5–10% of the total population was a Federal worker. That’s not of the labor force, that’s including children, retirees, students, and the unemployed. Enormous, long-term Federal investments do boost the local economies in which those investments occur, even if the taxes paid by other communities may weaken them.
Then in 1950, the Cherry Creek Dam was completed. And on that note, let’s talk about floods.
I can’t do justice to the history of flooding in Denver. Luckily, someone else has done the job for me. This article from a year ago catalogs the extraordinary history of floods around Denver, going back to Native American warnings to the first settlers that, seriously, white people, these rivers sometimes have very big floods. The short version is that Cherry Creek, which runs through Denver, flooded a half-dozen times from 1860–1920, killing dozens of people. This led to the construction of a dam, which collapsed in 1933. So the Army Corps of Engineers came back and built a bigger dam, which was completed in 1950. This choice saved innumerable lives.
After 1933, there wasn’t another big flood for a generation. The community got to thinking floods weren’t a big deal. As I noted, the city transitioned from a vice-and-services town to a town with more major industries and serious Federal investment.
But in 1965, a thousand-year-flood hit. This time, instead of Cherry Creek flooding, the South Platte River flooded. It essentially destroyed many of the industrial areas and poor neighborhoods along its banks, and destroyed more than a dozen bridges. Several people died, and 27 counties in Colorado were declared disaster zones. Tens of thousands of people lost their homes. The Cherry Creek Dam, however, held, despite the reservoir rising 16 feet in one day. Had the dam not been there, that 16-foot wall of water would have added to the 20-foot wall of water that already did hit the city, and it would have rolled right smack-dab through the middle of the city.
As of 1965, Denver’s core had been experiencing population growth since 1945. There were problems, of course: crime, worsening air quality, rising congestion. The suburbs had long-since become more popular than the urban core; that trend began in the 1930s but really heated up in the 50s and 50s. After desegregation in the 1960s, this ongoing suburbanization accelerated, and became more explicitly racialized. More car commuters meant that air quality got steadily worse.
1965, however, was a watershed year. Some suburban/exurban population weakness was compounded by the flood, and the city saw a population decline. From 1964 to 1966, the suburban/exurban population fell by 20,000 people, while the urban population fell by 10,000 people.
The Deluge came to Denver. But the real story I want to tell is what happened next.
Breaking Free of Path Dependence
What makes a bad neighborhood bad? Is it that the people there are bad? Is it that the neighborhood is systematically neglected by employers and governments? Is it some deficiency in the physical infrastructure? Is it some lack of imagination on the part of the whole surrounding community?
The truth is, we don’t totally know the answer to these questions. What we do know is that improving the quality of a neighborhood while maintaining a given population is very difficult. Conveniently, in Denver, right after the flood they tried two substantially different means of promoting recovery. One of these methods delivered a single impressive accomplishment, but at serious cost, and ultimately it failed to reach all its goals. The other method has shown much greater staying power and had many positive effects, while burning fewer bridges, but has not yet delivered as big a “ribbon-cutting moment.”
The first proposed recovery strategy was simple: bulldoze the dilapidated neighborhoods and replace them with gleaming new complexes of museums, parks, universities, etc. A group of city planners and other experts was gathered together to produce a plan to rebuild the impacted neighborhoods into something better than they were before. One town, Auraria, was indeed bulldozed, with over 10,000 people forced from their homes (turns out they weren’t quite as destroyed and dilapidated as planners had claimed), creating a displaced population, and a sour taste in the public’s mouth with regards to post-flood redevelopment. Having the government roll in and deliver the killing blow to a community knocked down by a thousand-year-flood is a recipe for bitterness.
In place of the lower-income community of Auraria, a huge urban campus was built for three different universities, that today provide an education for 50,000 students or more, and employ more than 10,000 people. The hospital treats hundreds of thousands of patients each year.
In simple terms, the Auraria campus is vastly more beneficial for the economy than, and probably houses nearly as many people as, the original Auraria neighborhoods. So it’s successful, right? These schools provide a key nexus of innovation and knowledge-generation, not to mention workforce training, that helped to propel Denver into a hub for more and more value-added industries, right?
Well, sure. But I actually think this is the redevelopment plan that was less successful. This top-down approach ended up incomplete: the Auraria Campus was the only of more than thirty projects that ever got completed, and it did so at the cost of a dubious ethical choice to toss thousands of people out of their homes, along with a galvanized anti-development movement. Part of the continuing population weakness of Denver’s core during the 1970s is in fact the result of the negative shock to housing supply of the flood and the demolition of thousands of housing units in Auraria.
Eventually, this top-down approach sputtered and died, mostly because it was incredibly expensive ($650 million price tag for everything to be completed).
But that wasn’t the end.
In 1964, the South Platte River was, for all intents and purposes, an industrial spillway. Today, it is lined with dozens of parks, and a dam upriver controls flooding. Notably, the Cherry Creek and South Platte River dams also now host the two most popular state parks in Colorado. This all happened because a smaller sliver of Federal money, under $2 million initially, was used to canvas residents near the South Platte River and get ideas for improvements they might actually like. Eventually, a project was launched to build a riverfront park, then eventually another, then some trails, then more parks. Near-river locations started to become popular for stadiums and businesses.
This approach of smaller-scale projects that actually listened to what residents wanted seems to have produced less negative pushback, while also being far cheaper ($130 million in total over the last 40 years).
In the end, both of these efforts helped the Denver community take vital steps towards making a more modern city. The downtown area got major university campuses, albeit at the expense of a historic neighborhood. The river area has, over time, been systematically improved and beautified, major runoff control measures have been implemented, and a boom in business investment amounting to over $13 billion has taken place. This investment has helped downtown Denver develop into the amenities- and employment-hub that many downtowns are becoming, providing vital services and jobs to their suburban populations.
Without the 1965 flood to shock the city into action and propel reconstruction, redesign, and large-scale rethinking, it seems likely that Denver’s downtown would have taken a quite different, and possibly worse, developmental path. As the Westword article linked above notes, a wave of historic preservationism struck the city not long after the flood, creating crippling zoning restrictions in many areas. Without the flood to catalyze reconstruction and new development, its possible that Denver’s ability to build for the future would have been set back by decades.
What is Denver’s Economy Like?
But it must be said: all these beneficial developments for investment, beautification, education, and employment are still population displacement. Stadiums take space. Businesses replace housing. Green spaces proliferate and end up restricting the housing supply too, driving up prices. Downtown Denver’s population kept falling for a long time, especially with the ebb and flow of oil and gas prices. Whereas once Denver was a silver town, now it’s an oil town. And that persists to this day, with a huge amount of fracking revenue ultimately making its way to Colorado. I’ve written before about the linkage between fracking areas and connected-but-distant metro areas like Minneapolis. It did not occur to me at the time that Denver might be linked, but it seems plausible that a large amount of the profits from the fracking boom end up making their way to and through Denver.
But is that really what’s going on?
Luckily, we can track the industry composition of Denver’s economy.
And, voila! The chart above shows the percentage change in each of several major sectors’ share of total employment in Denver. I only show sectors with large changes since the first year of data, and all data is from the BEA.
As can be clearly seen, the mining share of workers is way up since the first year where I have data in 2008. It’s easily the fastest grower. The information, Federal, construction, manufacturing, and farm sectors, meanwhile, are down. But paired with the rise in mining is a rise in employment in management; these may reflect large commodity or mining firms with managers in Denver. We also see, as in many cities, the growing prominence of healthcare and education.
Now, for a sense of scale, here’s the industry composition of Denver’s employment over the last 5 years, versus the national metro-area average:
The above chart is organized with the leftmost industries being those where Denver has an outsized share employed, and the rightmost where Denver’s sector is smaller than most metro areas.
It turns out, Denver is unique in that it has large professional services, financial services, and information sectors. Now, the professional services likely relate to large research universities, while the financial services almost certainly relates to Western Union being headquartered in Denver. And, by the way, Western Union was founded in Rochester, New York. I couldn’t find the story on when and why Western Union moved, but I can’t help but wonder if it relates to Western Union’s 1870s decision to make Denver their hub for the mountain west due to transportation access. Centuries-old logistics decisions could be impacting Denver’s employment today. The finance and insurance segment also includes many firms that work in commodities trading, which, again, is related to Denver’s status as a mining town. Then, of course, for information, we have historic strength but, as I showed, information is a rapidly shrinking share of Denver’s economy.
After that we have diminutive mining. Just 1.5% of Denver’s employment from 2009–2014, that was nonetheless nearly 3 times as prominent as the 0.6% in most cities. And much of the Management sector, also about 1.6% of Denver’s employment, versus 1.3% nationally, is also likely connected to mining.
On the other hand, we can see that Denver actually has a much smaller manufacturing, healthcare, retail, and state and local government sector than most cities. Manufacturing is perhaps no great surprise, and the small healthcare sector likely relates to a fairly young and healthy population. But the small state and local government share is genuinely surprising: Denver is a state capital! Why is the government share so much smaller than is typical?
Of course, employment isn’t everything. We can also see in which industries Denver’s wages are highest. Are there sectors where employees in Denver make lots more money than compared to the nation on the whole?
To begin with, a caveat: I have not controlled for occupational mix here. People working in “Mining” in Denver might have different job functions than people working in “Mining” in Fargo. But this does at least give us an idea of some structural pieces of Denver’s economy.
First of all, Denver’s wage rate for most sectors is pretty close to the national average. But it is exceptionally high in management. Again, I strongly suspect that the “management” being captured here is management of oil- and commodity-related firms. Meanwhile, we see that other exceptional industries in Denver like mining and information also have higher-than-average wage rates. All of this comes together to suggest that Denver may be a good place for these industries, where workers are more productive, so can reasonably be paid more.
Then we have the other side: finance and insurance. This distinctive sector seems to be located in Denver because Denver is a cheaper place for the work than most finance and insurance hubs. This may also be a firm- or occupation-specific effect: maybe the people working in finance in Denver have atypical roles versus people in finance around the nation. But then again, many of these folks are commodities-related traders, or working in a major financial firm like Western Union, so I’m not sure how credible that claim would be.
Denver Today
As the first chart shows, Denver’s city center and suburbs have both shown strong growth for at least 25 years, since the oil crunch in the early 80s. The city has gradually gotten a handle on air pollution, crime has fallen, and its reputation for vice and corruption from the early 20th century is pretty thoroughly expunged. Today, Denver, like most Colorado cities, draws a mix of young and old migrants with promises of a high quality of life. This quality may arise from hip urban areas, nearby natural amenities, a nice climate, affordability, or good job growth. Different people will value different things.
But like most successful cities, the striking thing about Denver is not its One Big Thing, it’s that it simply doesn’t have any crippling weaknesses. When the fracking boom busts, Denver will have some struggles, but its other sectors seem robust enough that such a bust shouldn’t be as crippling as earlier commodity cycles have been. Crime is down. The smog problem of the middle of the century has mostly been addressed. Devastating floods have mostly been controlled. In the huge 2013 floods, less than 20,000 houses were damaged and under 2,000 destroyed, and flood control continues to improve. In Denver itself, the damage was particularly well-controlled, with the worst impact felt in Boulder.
However, Denver faces a new menace: the menace of insufficient housing.
The above chart shows new housing permits in the Denver MSA versus population increase. As you can see, the two tend to go together. But, critically, housing permitting in Denver fell dramatically from 2005 to 2010 even as population growth accelerated.
Now, for what it’s worth, Denver’s urban core has done better than the suburbs. While the core has build 0.54 new housing units per additional resident in the core since 1980, the suburbs and exurbs have built about 0.43 housing units per additional resident. But then again, the urban core has an older housing stock on average, so probably had more housing depreciate or need to be demolished, so maybe that 0.54 overstates actual increases in housing supply. But, then agaaaaiiin, the 1965 flood should have destroyed a lot of the worst housing stock, so maybe there isn’t that much older housing? Who knows!
But here’s the thing: Denver on the whole added 0.44 housing units per additional resident. If you recall my Phoenix post, that means Denver has added more housing supply compared to population than Phoenix, San Francisco, San Diego, or San Antonio. Go Denver, right?
Well, maybe. But if we change our sample to supply added in the last 5 years versus population added in the last 5 years, the story changes a bit.
Now, every city in my (admittedly small) sample has seen a big decline in permitting versus population growth. And Denver remains at the top tier of permitting rates, as it has for a while. Denver has traditionally been pretty friendly to new construction, with some exceptions.
But the first glance showing, “Oh, everybody’s had declines!” would be wrong.
Since 2010, Denver’s average 5-year permitting ratio has been 0.198, the second highest in my sample behind Dallas.
But alas, that represents a 39.4 percentage point decline from its pre-2010 average, of about 59.1%. That is the largest decline in my sample. Represented another way, the permitting ratio has fallen 66.6% in Denver, the 2nd steepest rate of decline outside of San Diego, where the permitting rate has declined 68%.
All that to say: Denver’s housing supply is tightening a bit. Sadly Denver’s definition has changed a gazillion times so the ACS files on it are inconsistent and hard to compare, so I can’t easily show changes in rental costs.
And, look, maybe housing construction will heat up in Denver soon. If rents do go up, that can be a good signal to builders to build more. And maybe Denver hasn’t really tightened up any key policies in the last 5 years or so, so new development will be no big deal. Maybe. I hope so.
We’ll see.
Check out my Podcast about the history of American migration.
If you like this post and want to see more research like it, I’d love for you to share it on Twitter or Facebook. Or, just as valuable for me, you can click the recommend button at the bottom of the page. Thanks!
Follow me on Twitter to keep up with what I’m writing and reading. Follow my Medium Collection at In a State of Migration if you want updates when I write new posts. And if you’re writing about migration too, feel free to submit a post to the collection!
I’m a graduate of the George Washington University’s Elliott School with an MA in International Trade and Investment Policy, and an economist at USDA’s Foreign Agricultural Service. I like to learn about migration, the cotton industry, airplanes, trade policy, space, Africa, and faith. I’m married to a kickass Kentucky woman named Ruth.
My posts are not endorsed by and do not in any way represent the opinions of the United States government or any branch, department, agency, or division of it. My writing represents exclusively my own opinions. I did not receive any financial support or remuneration from any party for this research. More’s the pity.
