This post on Best Of A Great Lot is a part of a series on the subject of designing a new form of governance. Each piece aims to stand alone, but fits together on the Table of Contents.
In our market economy, some of the services that are invented by private actors become infrastructure. The last century has seen a vast philosophical and political gulf among different ideologies in terms of whether that infrastructure should be managed by the public (nationalization) or by privately run companies (privatization) and with private companies, what kind and degree of rules (regulation) those companies should be placed under. Because there are major flaws in both directions, we're always disappointed. I aim to lay out the flaws and strengths of each model in pursuit of designing a better path.
Philosophically, the extremes of this spectrum are held by communism and libertarianism. The core argument for communism is that private ownership is the root of disparity and inequity. At the other end, the argument for libertarianism is that the exercise of governmental power exists to reduce freedom.
The modern liberal democratic compromise is that many pieces of infrastructure will be privately managed but with varying degrees of regulation. In most developed democracies, some aspects of infrastructure are still or newly governmentally managed (nationalized), but there is no clearly demarcated line between which things are nationalized and which are privatized.
In brief:
Capitalism generates new infrastructure
People distrust for-profit infrastructure.
Nationalized infrastructure often becomes inept or corrupt.
Neither nationalized nor privatized is ideal, so there is an ongoing tug-of-war in both directions.
The compromise position is privatized with regulation.
Regulation prevents the worst excesses, but also creates barriers to entry and to change which prevent innovation.
Capitalism generates new infrastructure.
Once we lived without electricity and running water. Now they are ubiquitous. Once we had never heard of the internet, now most people use it. Once, few people had a bank account, now most do. As technology advances and people imagine new ideas, capitalism allows entrepreneurs to turn some of these new ideas into profitable industries. Some of these industries are especially profitable because everyone wants them, and as we come to depend upon them, they become infrastructure. Once they are infrastructure, other entrepreneurs can build new services which assume that we will have the old ones. Our dependence on them thus shapes society. I wrote more about infrastructure here.
People distrust for-profit infrastructure.
Many people deeply distrust for-profit companies running the infrastructure we depend upon. You see this distrust in the continuing appeal of communism despite its abject failures in reality; in every dystopian science fiction novel that imagines corporations replacing governments; in the run of the mill demands from citizens that corporations be more regulated; and many other ways.
There are plenty of solid reasons to distrust for-profit infrastructure. A fundamental truth of infrastructure is that we depend upon it, and so if we are to have an equitable society, we must distribute infrastructure equitably. For-profit companies have little incentive toward delivering infrastructure equitably; among other reasons, richer populations are more profitable.
Worse, for-profit companies aren't just meant to turn a profit: they have investors actively demanding that they grow that profit over time. But infrastructure is often a bad place to continuously grow profit simply by delivering a service well. Instead, investors have to find other ways to squeeze growth out, often at the long-term expense of society.
Consider Overdrive, which makes software that libraries use to give their patrons access to ebooks. Overdrive was recently bought by KKR, a notorious Private Equity firm. Over the last 3 decades, Overdrive went from a tiny, brand new company to being the software provider used by 95% of libraries in the US. Once you've sold to 95% of your potential customers, and every library patron in the country uses your software on a daily basis, the options for increasing growth are: build something new; raise prices; or squeeze costs. The first one tends to be risky, as organizations that are designed to exploit a niche tend to be bad at exploring a different niche. Private Equity also tends not to be interested in funding risky exploration. Instead, they prefer to go for raising prices or reducing costs. But now that we depend upon Overdrive to provide a significant aspect (software) of a public service (libraries), the model of raising prices and squeezing costs is only going to make life worse for library patrons. If Overdrive goes the way of Toys R Us (also a KKR acquisition), we're all worse off.
Some of the problematic dynamic here is monopoly power. KKR couldn't shaft library users nearly as easily if there were 3, 4 or more software providers for libraries. But some of it is simply that as a product becomes infrastructure, it matures into a situation where profit growth is at odds with the needs of society. There's a limit to the size of the library software market. Growing profits once you're in a mature market generally requires doing something that's bad for users of the infrastructure: worsening quality, raising prices, using society's dependency as leverage to sell things that we don't need through bundling, jumpstarting into some more profitable market and deprecating the infrastructure entirely, etc. We've seen Comcast try every version of this with its internet service duopoly.
But infrastructure providers love to exploit monopoly rents, because the barriers to entry into the market (regulatory, financial, dependencies from other organizations, etc) make for a great place to make money on shitty performance.
Consider planned obsolescence in mature consumer goods. For many technologies that we've had around for a hundred years - appliances, for example - we know quite well how to make them last. There is little real innovation happening in many of these. Laundry machines and kitchen appliances from fifty years ago operate very similarly to today, and many of the fifty year old models are likely to outlast a brand new one. There have been some significant changes - especially efficiency improvements - which have largely been driven by governmental regulation more than the market doing its job as competitive driver of innovation. Then there are a few changes that are genuinely innovations, like electric induction in stoves. But for 90% of consumer appliances, the fundamentals are the same as they were 50+ years ago.
Yet despite the fact that the vast majority of appliance tech is unchanged and we know how to engineer for resilience and for repairability, these objects are designed to be minimally repairable and last 10-12 years. Appliance companies need you to buy new models if they're going to justify their stock prices. They have to grow forever, even though it's not good for society for them to do so.
At the heart of the justification for why capitalism is good1 is the claim that it's a system whereby entrepreneurs are incentivized to satisfy customers’ desires. This incentive to satisfy desires is built on top of free choice and the possibility of profit. As a society, we’re capable of recognizing that a coerced action doesn't count as a free choice and that the level of freedom of a choice depends on the level of coercion that's applied. Infrastructure is a case where there's a spectrum from entirely optional (which probably isn't quite infrastructure yet) to something we're so dependent on that we have little true choice.
It's a rare person who opts out of housing, electricity, cooking appliances, cars2, the internet, nursing care, or other things we can safely call infrastructure. Many of the people who do so are effectively opting out of society entirely. Some of these services are still offered by many competing companies. But for more and more, the number of options has been rapidly consolidating.3 In the case of monopolistic infrastructure, the dynamics of the free market start to resemble the dynamics of drug dealing. Stringer Bell, the drug kingpin from the TV show The Wire captured these dynamics really well:
I know; sh*t *is* weak, but, y'know, sh*t is weak all over. The thing is, no matter what we call heroin, it's gonna get sold. Sh*t is *strong*, we gonna sell it; sh*t is *weak*, we gonna sell twice as much. You know why? 'Cause a fiend, he gonna chase that sh*t no matter what. It's crazy, you know. We do worse, and we get paid more. The govermnent do better, and it don't mean no nevermind. [pointing to the money] This sh*t right here, D, is forever.
Of course we distrust companies selling us infrastructure. We don’t have a lot of choice about buying, and we know they know that. Active competition can help this situation, but it only goes so far. Of course they’re going to find ways to profit from that situation and not in ways that align with our interests. This is not the free market that the boosters of capitalism imagine.
Nationalized infrastructure often becomes inept or corrupt.
At the local level, everyone loves to complain about the DMV and the Building Permit office. There are lots of government run institutions that at least one side recognizes are terribly run, like liberals and the police department, or conservatives and the various agencies that we pay taxes to. What’s true of the local level is true at the national level, even if our perception of it is largely warped by politics.
At the heart of the problems with nationalized infrastructure is a key fact within every bureaucracy: every attempt to cut waste or innovate requires political capital to do so. The people who are running the waste usually want to keep it - either they have bought into some promise that it will help in the long run, or they're focused on local incentives like increasing their budget and headcount. There's always a leader who is bought into any particular piece of waste, and that leader has some degree of political capital within the organization. It's certainly possible to get a wasteful project cancelled, but doing so takes bravery and connections. When there's no direct financial incentive to the leaders of the organization, they're vastly less likely to pay attention to the waste. Instead, any waste they can reasonably justify is evidence of the need for bigger budgets for them.
On the efficiency side, the economic theory of privatization is that private companies have a stronger incentive to reduce waste that publicly run organizations can never match. The CEO and board who are compensated based on share price are much more likely to cut costs to boost shares. When they cut the long-term health of the company to earn a quarterly profit, they do damage to the company. But when they cut unnecessary initiatives and other forms of waste, they reduce the cost to deliver the services that the company provides. When a similar service is provided by the government, there's nobody who earns money by cutting that waste.
Competition can drive that efficiency deeper into the organization. If there is active competition, the competitor who offers cheaper, better service should be able to win out against the competitor who is less efficient and more expensive.4 Governmental organizations tend not to be offering their services within a competitive marketplace.
On the corruption side, there are always people who salivate over the chance to turn something publicly run into their own personal moneymaking fief, or who know someone outside who can be paid to do whatever they can justify.
We could argue endlessly about whether the specific problem within our society is more driven by bureaucratic inefficiency or corruption. For my purposes, it doesn't matter: we know that both exist, and we know that they prevent nationalized infrastructure from being efficient. So we have to find a path that doesn't encourage them.
Neither nationalized nor privatized is ideal, so there is an ongoing battle in both directions.
Most ideologues can point to why clearly one or the other path is better in their particular domain of caring. Maybe they believe that we should nationalize healthcare, because "people shouldn't earn profits off other people's health" or they think we should privatize the post office or road maintenance or tax collection because clearly FedEx proves that privately delivered mail is fine.5
Each side has downsides. When a particular aspect of infrastructure is run privately, we feel the pain from the privatization path. When it's run nationally, we feel the pain from the nationalization path. So we're always ready to argue about switching any particular piece of infrastructure to the other direction. The grass is simply always greener in our imagination.
The compromise position is privatized with regulation.
Apart from corrupt options, there’s not a lot of proposals out there to let agencies profit off of running more efficiently.
However, we have found some ways of reducing some of the negative externalities that privatized infrastructure tends to produce - primarily by regulating those industries. Through this we can continue to have some of that personal financial incentive for those who run the organizations, but we can try to reduce the inequity and other problems they would otherwise produce.
Unfortunately, regulation has its own problems. Regulated industries have significantly less competition, and as a result, innovation. They're harder and more expensive to break into, which reduces the number of entrepreneurs available to work on them. One longstanding complaint with Silicon Valley has been that investors and entrepreneurs there focus primarily on apps and other things that don't matter to the day-to-day functioning of society. Why do they invest in apps and not more important things? The economics of the situation have driven them to. To build a new app and prove its value might cost a few million dollars. To build something new in a regulated industry like healthcare costs easily ten, twenty, a hundred times as much. If healthcare profits were ten times as high as app profits, we might see investors taking that bet, but since they're often not, regulated industries are often just a bad bet for an investor.
Worse, regulation often slows or stifles innovation even when entrepreneurs and investors line up. Great regulation that is flexible to the changing realities of our world is absolutely possible, but it's much harder to write than bad regulation that simply enforces what we currently know.
Consider nuclear power. Our current regulatory environment was built in the 1970s and 1980s along with the nuclear power plants we built. Not surprisingly, the regulations are designed around the design of those power plants. New plant designs that are fundamentally safer have a tough time getting regulatory approval because they don't include the same safety engineering: the design simply doesn't require that engineering to be safe. It's possible for regulators to update based on the new designs, but it's hard, and it adds another layer of complexity to a company attempting to innovate. So it's not terribly surprising that we've seen most of the innovation in nuclear power plant design happen with an aim to build things in China.
Where does this leave us?
None of these paths is effective to the level that we're aiming for. We need to design an ecosystem that allows entrepreneurs to innovate new infrastructure, encourages stable and equitable infrastructure delivery, and prevents mature businesses from either running old infrastructure rapaciously. It can be done, we just have to keep the incentives in mind as we build this design.
If you find my work interesting, please share and subscribe. It helps tremendously.
I don’t intend to spend much time defending capitalism, but while reading for this article I did run across a particularly great quote that can do the job. From the appropriately titled In Defense of Global Capitalism:
The growth of world prosperity is not a ‘‘miracle’’ or any of the other mystifying terms we customarily apply to countries that have succeeded economically and socially. Schools are not built, nor are incomes generated, by sheer luck, like a bolt from the blue. These things happen when people begin to think along new lines and work hard to bring their ideas to fruition. But people do that everywhere, and there is no reason why certain people in certain places during certain periods in history should be intrinsically smarter or more capable than others. What makes the difference is whether the environment permits and encourages ideas and work, or instead puts obstacles in their way. That depends on whether people are free to explore their way ahead, to own property, to invest for the long term, to conclude private agreements, and to trade with others. In short, it depends on whether or not the countries have capitalism. In the affluent world we have had capitalism in one form or another for a couple of centuries. That is how the countries of the West became ‘‘the affluent world.’’ Capitalism has given people both the liberty and the incentive to create, produce, and trade, thereby generating prosperity.
Except of course in NYC and the other rare city whose public transportation is sufficient to allow people to opt out. Everywhere else in the US, cars are essential infrastructure since they are the only way to get to work and the grocery store, among other essential places.
BIG, by Matt Stoller, is a newsletter entirely dedicated to documenting the problems of monopolies on American life.
This only holds true in markets where the savings can actually be passed on to the consumer. Many buyers, such as corporate and governmental buyers, are much less price conscious than worried about making a decision for which they can be blamed. Thus the slogan "Nobody got fired for buying IBM." There are also lots of other cases where competition does not seem to drive prices down, such as college tuition and healthcare.
The word clearly is one of those words to watch out for, as it seems to regularly cover a complete lack of solid evidence.