Government office buildings are apparently more about status than I thought
So It Turns Out, #50
Hi,
So here are some things I’ve been thinking about this week.
Government office buildings are apparently more about status than I thought
When we think about the accomplishments of past civilizations, we often think of their buildings used for government (e.g., palaces, castles, hunting lodges). Lots of old stone buildings, basically, that signal “We considered ourselves to be a powerful culture” to visiting barbarians then, and tourists now. (Same thing?)
Here’s the weird thing: aside from a few brief periods in late Roman history1, many of those historic sites in the Western world2 are, well, homes. Feudalism, by its nature, had much closer linkages between leaders and governance than we’d see in modern times, so why would they bother to go anywhere else? When you read accounts of where medieval European courts, they consisted, basically, of lords either at their own manors/castles/palaces, or going on month-long vacations to their vassals’ manors/castles/palaces and holding court there.
This probably makes sense in an ancient context. After all, when leadership is much more personal, what would be the point of a separate building to work in? And if you want a change of scenery, why not impose on a vassal — after all, isn’t that what they’re for? The structures of power are the structures of awe.
But it’s not what we do in the modern era, is it?
You can, indeed, still see some traces of this in most modern democracies — the White House in America, Number 10 in London, etc. — but I’d venture that most of this is a sort of post-nobility overhang on the West of that the offices of the leader are where the leader lives, rather than a considered decision about whether it’s healthy for the leader of the free world to have no cognitive separation between life-and-death decisions and relaxation. Rather, the assumption seems to generally be, and has been for most of the time, that everyday government employees work from a place of official duty provided by their employer (e.g., an office building), barring limited counterexamples.
Of course, this offers many salutary advantages for governments to mitigate their risks and enable their capabilities.
On the mitigation side, it’s fairly obvious that modern administrative, surveillance, and bureaucratic states rely on piling up all sorts of sensitive data at a scale beyond that of a single trusted Walsingham.3 It’s fairly obvious why there are all kinds of government work in modern states that rely on giving a fairly expansive number of individuals access to fairly sensitive data, beyond the scale at which a more personalized model of “I will kill my vassal if he betrays me,” is sufficient. And much of that data either simply cannot (e.g., classified work) or really really shouldn’t (e.g., health data work) be routinely done outside of government-controlled buildings. Indeed, the data in the GAO report indicates that agencies with more work of that nature, generally speaking, are clustered in the top quartile of return-to-office rates.
The capability side is perhaps less obvious. After all, casual inspection of the (originally government-grant-funded) Internet that you’re using to read this letter will remind you that governments are uniquely good at building long-distance communications systems. Indeed, I actually think being in person is relatively less of an enabler for public-sector organizations than private-sector companies (e.g., arguably the first power user of telegraphs for government purposes was Abraham Lincoln)4.
Nonetheless, many government officials do believe being in person is important. This is probably driven by three tendencies:
“Everyone knows” that government employees don’t work hard. Suffice it to say that this letter strenuously disagrees with that diagnosis as contemptuous of many government employees who routinely forgo sleep to keep you safe from nuclear war, but admits that as government employees are humans, many of them are certainly likely to benefit from closer supervision.
Senior government officials tend to believe that “being close to the flag” is an important way of knowing what is going on — for example, I’ve had clients at many government agencies over the years whose official (relatively more) snazzy offices are in building X, but their boss is in building Y, so they spend all their days working from a crash space in building Y. This tendency percolates all the way down the hierarchy. (Government is, perhaps, far more personal than the bureaucratic abstraction layer we draw over it wants to admit.)
Finally, probably some of this was sheer inertia — that’s what people did for a long time, why rock the boat with major changes? Sure, there were some experiments with work from home and alternate schedules, but why risk one’s career on proposing something more expansive?
So what happens when you’ve told a huge number of those employees that they should work from home and then attempt to reverse that decision?
Earlier this month, the GAO published a report on the Federal Government’s utilization of its office buildings, finding:
Seventeen of the 24 federal agencies in GAO’s review used an estimated average 25 percent or less of their headquarters buildings’ capacity in a three-week sample period across January, February, and March of 2023. On the higher range, agencies used an estimated 39 to 49 percent of the capacity of their headquarters on average.
There are some important caveats to that bottom-line finding, which we can generally bucket under our favorite category, “Have you ever tried seeing like a state? It’s hard”:
The GAO report admits that the data is extremely inconsistently-collected5 and gathered for administrative headquarters buildings only. As a result, the report (as far as I can tell) doesn’t have the ability to say, “well, yes, Alice wasn’t in the DoD’s administrative headquarters, which is officially Alexandria, VA, Mark Center on Monday. Because she was in meetings all day in the Pentagon, which apparently not the administrative headquarters of the Department of Defense and thus literally not in the data set for this report. Ignore that every human ever working in the Pentagon has complained about how crowded it is there.”67
The report openly admits that there is no known target for capacity utilization of government office space that has ever been established by anyone (The GAO notes, “For example, we calculated for one of the headquarters in the lowest use quartile that if all assigned staff entered the building on a single day, it would still only use 67 percent of the building’s capacity based on its usable square feet”). Ironically, the GSA, who is responsible for managing the buildings, has among the worst utilization metrics.8
GAO gathered data in 3 months that generally have quite a few employees taking winter weeklong vacations that abut the long weekends caused by federal holidays, and apparently didn’t control for that at all.
The report is specifically from a time period where DC had a COVID spike, and many people I know reverted to more COVID-cautious behavior (or had COVID; I did).
Agencies don’t agree with GAO on what counts as office space — for example, apparently one agency very politely noted that they felt confident that their cafeteria shouldn’t be counted in the denominator of total office space.
There’s no standardized way of tracking how hard it is to renovate an existing building — for example, the Department of Treasury all-but-told GAO, “Astonishingly, it is expensive to renovate our historically-conserved classical building that is in direct sniper line-of-sight to the President of the United States, so we have fallen behind on that sweet hot-desking plan you suggested.”
And, oh yeah, exactly zero percent of this report is devoted to the question of whether or not government employees perform better or worse when given the choice of where to work from…
Still, there is strong reason to believe the report is directionally accurate; other ways of triangulating the same work-from-home behaviors, such as WMATA subway fares, are also down in that time period. We’re, indeed, not using lots of office spaces that we have.
And this costs the American taxpayer real money; the GAO says the GSA rents rents 180 million square feet of office space on behalf of federal agencies at a cost of some $5B, and that the government could plausibly repurpose even some of the 511 million square feet that it owns under GSA management and spends $2B to maintain annually.9 (In among the driest asides I’ve ever read in a government document, GAO notes, “In the local economy, unneeded federal properties and land could be put to productive use. For example, the private sector successfully converted an unneeded post office in Washington, D.C., into a hotel.”)
I do feel some genuine sympathy for those managing space on behalf of agencies at the multi-decade time scales of government (or, heck, just the time scales for building any sort of building in Washington these days). No doubt there are some no-regrets moves, but a lot of decisions here will be a bet about what the future will hold. And when betting about the future, governments prefer to hedge their bets whenever possible; that’s probably even more true when deferring a decision has long-term costs, but short-term savings.
After all, it’s probably equally plausible to believe that a few years from now, we’ll be still in a) the current status quo, b) a return to pre-COVID norms, or c) a dramatic slashing of presence in the DC area to move billets out to other states (as Trump has, indeed, suggested he will do if returned to office). Why spend a lot money on making a big change that then you might not even use? Bureaucratic memory is long, and in a future global war or pandemic (especially the latter), no one will begrudge past spending that kept more office space around.
But another part of this is said by the GAO report, somewhat softly: government agencies really like having prestigious real estate, and don’t want to give it up to other agencies, and this pattern replicates all the way down the org chart (e.g., Department X’s Bureau A doesn’t want to give up space to Bureau B, and so forth). You can’t point to your employees’ homes when you’re giving your boss an argument for needing more resources, but you sure can point to the size of your department’s conference room.
I’ll admit a fact here, dear reader: while writing this letter, I didn’t quite know where I was going to conclude here.10 As someone who started his career in Washington crammed into a very small office funded by War on Terror money, I was curious to see how life in government offices had evolved in a work-from-home era. (And yes, this report doesn’t shed light on the defense and security world to the same degree as more open parts of our government, but it still does shed some.)
And so, the conclusion that I’m coming to is one that, well, I don’t really enjoy. It’s this: the Federal Government, to preserve many essential operations in the midst of a (hopefully) once-in-a-lifetime event, told its employees that it trusted them to run one of the largest democracies in human history in significant part from their bedrooms. And that turned out, in the cosmic scheme, surprisingly well for anyone other than the DC restaurant industry and those looking for an affordable two-bedroom rental inside the Beltway.
But humans are still humans, and it turns out that far more of the role of the government office building — yes, even the ugly ones with mildew — is to be a palace. Maybe not one that people live in any more, but one still intended to demonstrate and hold onto power and status for bureaucratic competitions. No one was really managing the buildings as resources, even when at a cost of literal billions of dollars, and we’re not willing to pony up the (in comparison far smaller) sums of money that we’d need to spend to make those buildings more useful and attractive to civil servants.
I don’t blame the average civil servant — they’re operating under the incentives we give them, incentives familiar to everyone in the private sector, too. And (as I hope to write about more at a later date), professional training for civil servants is shockingly lacking when it comes to procurement. But it puzzles me — even as someone who’s lived in DC for almost my entire adult life — how we can be so bad at these basics. We certainly can trust someone to make life-or-death decisions from their couch, but we apparently shouldn’t trust those very same people to sign a lease.
And I don’t know, entirely, how we fix that.
I’m looking for new opportunities; I’ve been especially successful in the past in corporate strategy and consulting roles where I work closely with, yes, public sector employees, but also technical, product and legal experts, and am looking for similar for my next role. I’m open to a range of different industries, but am especially interested in organizations where I can help keep people safe from bad guys, and/or help them get easier access to government services and benefits that they deserve. If you’re aware of an interesting opportunity, I’d love to chat with you and learn more.
Disclosures:
Views are my own and do not represent those of current or former clients, employers, or friends.
A complete discussion of the debates in ancient Greek and Roman history about to what extent bureaucracies really existed is outside the scope of this letter, but suffice it to say that generally the Roman empire was a far more indirect empire with low investment in administration than you would assume, at least until Diocletian. A good summary of the is here: https://acoup.blog/2022/01/28/collections-rome-decline-and-fall-part-ii-institutions/
A discussion of imperial bureaucracy in China is outside the scope of this letter, because I don’t have enough expertise in it to avoid bungling it. But it’s a big counterexample too.
Indeed, there is a credible argument that the ways early modern states beat Napoleon’s genius was by scaling up the general staff model to give their less-skilled generals the ability to not keep the whole problem in their own heads while fighting Napoleon.
Yes, yes, the Barack Obama-Blackberry comparison writes itself, we’re all clever for thinking of it.
The GAO notes, “Currently, each agency establishes its own measures and standards for office space utilization. We found that agencies use a mix of badge swipes, network logins, self-reporting, or guard tracking to measure attendance at their headquarters. These differences feed into additional differences in how agencies measure building capacity. Not all agencies agreed with our approach to measuring utilization because they use different metrics for office space planning. For example, some agencies attribute a certain square footage per staff person, while others count physical workspaces.”
Or, for that example, deployed to a US base in Bahrain or similar. Doesn’t count.
Ironically if you believe in-person presence matters, you might well expect lower utilization as measured — if in-person meetings matter, then whenever your staff go to important and useful in-person meetings at other agencies or even other offices for your own agency, they get subtracted from your utilization but probably not reflected in the other agencies’ metrics…
As discussed elsewhere in this letter, presumably some of that is due to GSA employees, y’know, going to other buildings to do their jobs instead of sitting at their desks twiddling their thumbs. You might well imagine the folks responsible for managing the other buildings to, well, go visit those other buildings often…
This seemed low to me at first glance; parsing the text, it seems that the lease number only includes leases done by GSA, not, e.g., an airbase overseas. Similarly, the costs of maintaining owned properties under GSA management probably isn’t inclusive of all buildings owned by the federal government.
Arguably the main reason to write an essay!