What Are Governments Actually For?
What is government for?
It’s a simple but important question. Yet, it almost never gets asked within politics.
The questions that are asked are more specific. Which party has a better healthcare plan? Should taxes be higher or lower? Should the government regulate this industry or subsidize that one?
Those questions are asked constantly, argued exhaustively, and resolved temporarily by elections, then reversed or altered in the next election.
The question that doesn’t get asked is the one underneath all of those. It is the one that would change the character of every argument downstream if it were answered honestly and held consistently.
What is the purpose of government? What are they really trying to achieve, as a first principle? What is the destination toward which the instrument is being pointed?
Most people, when asked directly, would give an answer that reflects their tribal map. Libertarians, socialists, republicans, liberals, they all have their answers, but these generally tell you what the government should do—or avoid doing—without telling you what it is for.
These are downstream answers masquerading as foundational ones.
And they often start backwards, starting from “what do I want?” then believe the government should supply them with it.
The absence of a genuine first principle produces a specific kind of political dysfunction. Not the obvious dysfunction of bad policy, though that follows, too. It’s a deeper dysfunction of an argument conducted at the wrong level of abstraction.
Two sides fighting about the how before they’ve agreed on the what. What follows is navigation of specific territories with increasingly complex maps, with no binding compass.
It would be as if the question “how do diseases work?” never got asked. What would follow is increasingly complicated and competing maps for “how do you stop a cough?” instead of increasingly complex maps at the right level of territory—namely, “how do diseases work?” The fractal of opposition is instead building its complexity in the wrong direction.
That’s not to say specific maps around coughing or fevers are irrelevant. When germ theory arrived, these maps gained a compass. The cough map became more accurate because it was now situated within a coherent account of what coughing is for. The fever map became more accurate because it was now situated within a coherent account of what fever does.
The first principle for government does the same work. It doesn’t replace the specific policy debates, but situates them within a directional compass.
Because, to put it bluntly, the only way to compare the accuracy of two maps is to first agree what the map is for.
This essay proposes that first principle. Not a complete political theory. Not a map of every policy question. A compass that gives us a unified direction against which the downstream arguments can be honestly tested.
Here it is: the government’s role is to maximize freedom for the greatest number of people.
Why This First Principle
The specific value of this formulation is that both the left and the right can genuinely agree with it without abandoning the core commitments of their already established maps.
The right already talks about government overreach restricting personal freedoms, the free market, individual choice. It already speaks about the nanny state of slow-moving, bureaucratic judgement. It wants to reduce taxation to let people decide for themselves how to direct their resources. These are all genuine observations about genuine restrictions built into the core intuition of the right, and the first principle captures it.
The left, in much the same way, talks about corporations with monopoly power over essential services. It worries about a labour market which restricts worker’s rights and freedoms. It acknowledges how poverty restricts upwards mobility or access to essential rights. It tries to address social freedoms, like the freedom to marry whomever you want, to love whoever you want. The left’s core political intuition also stems from genuine observations about genuine restrictions, and the first principle captures it.
This formulation, then, does something other political frameworks fail to do. It treats government overreach and corporate overreach as the same category of problem.
The factory that poisons the river is restricting the freedom of everyone downstream. The monopoly that controls access to essential infrastructure is restricting the freedom of everyone who needs it. The regulatory process that makes housing construction effectively impossible is restricting the freedom of everyone who cannot afford a home.
Once the first principle is established, the argument between left and right changes character. It is no longer about whether freedom matters. It’s clear that both sides believe it does. What matters is about which actions expand rather than contracts the overall freedom available to the most amount of people.
That is a harder and more honest argument than the current one. It cannot be resolved by each tribal map alone, but requires some combination of both/neither as we partake in a genuine investigation of the territory.
The Problem With Freedom’s Complexity
Stating the first principle honestly requires immediately acknowledging its central complication: freedom is not a single measurable quantity. It is a family of related but distinct values that don’t always move together, and are sometimes competing.
Isaiah Berlin identified sixty years ago two types of freedom: negative and positive freedom. Negative freedom is the absence of constraint, the protection from interference, and the right to be left alone. Positive freedom, on the other hand, is the freedom to. It is the actual capacity to act, to pursue the life you have reason to value.
The person born into poverty has extensive negative freedom in the strictest sense. They are not literally prevented from starting a business, getting an education, or achieving upwards mobility. It is their positive freedom, their ability to do these things, that is near zero. As a result, things like a progressive tax system, social housing, and public schools, may be used to expand their positive freedom, while restricting the negative freedom of others. Both of those are freedom effects, but the first principle does not resolve which freedom takes priority when they trade off against each other. It only says to maximize freedom for the greatest number of people.
This is not a defect in the first principle. It is the honest condition of having a compass before you start to build a map. The compass—the first principle—simply ensures the argument is being conducted at the right level of analysis. Not “should we have regulation or not?” But “does this regulation expand or contract the real freedom available to the most people, when both the benefits and the costs are honestly accounted for?
That question cannot be answered by tribal affiliation. It requires genuine investigation.
The Market Has Always Known What It’s For
There is one institution in modern society that has never suffered from the hypothesis problem: the free market.
We do not consistently see new documents about its values or direction because its hypothesis is embedded in every transaction, every quarterly report, every hiring direction, and every product roadmap. The hypothesis is stated, measurable, and relentlessly tested in real time.
The purpose of the free market is to extract maximum value.
This is why markets work as efficiently as they do. Not because of the intelligence, virtues, or capabilities of private actors. But because the compass is unambiguous and the feedback mechanism is immediate and brutal. The company that drifts from its hypothesis, which optimizes for something other than value, gets corrected by the market before the drift becomes terminal—or they are terminated.
A car company’s purpose is to make the best car it can at a price people will pay. Everything good that follows—the engineering innovation, the manufacturing efficiency, the customer experience—is a byproduct of that single clearly stated mission. Nobody debates what a car company is for. Nobody needs to. The hypothesis states itself in the price, the sale, the profit, the loss. The market’s genius is that it makes the hypothesis unavoidable.
This clarity is also the precise source of every market failure worth naming. The river gets poisoned because the river is not in the hypothesis. The worker gets exploited because worker dignity is not in the hypothesis. The housing doesn’t get built where people need it because need and profitability are not the same variable. The drug that treats a rare disease doesn’t get developed because rare diseases don’t extract maximum value. The neighbourhood gets hollowed out because the hypothesis said move the factory to where labour is cheaper and it was right.
These are not market failures in the technical sense. They are market successes. The market did exactly what its hypothesis specified. The hypothesis simply didn’t include the river, the worker, the family who cannot afford a home, or the patient with the rare disease. So they weren’t included in the outcome.
This is the moment where government’s role becomes precisely legible.
Not to replace the market’s hypothesis, but to balance it with another hypothesis the market cannot structurally hold, then intervene specifically where the market and the government diverge.
Labour law. Environmental regulation. Public education. Universal healthcare. Each of these is the government holding the hypothesis the market cannot hold—the freedom of the worker, the future of the environment, the capability of the child, the health of the person—and intervening at the specific point where the market’s relentless value-extraction diverges from the maximization of freedom for the most people.
The relationship is not adversarial. It is complementary in the precise sense. The market is the most powerful hypothesis-testing machine ever built for the domain its hypothesis covers. Government is the institution with the democratic mandate to hold the hypothesis the market’s domain excludes. Together, with both hypotheses clearly stated and honestly pursued, they form something closer to the complete investigation than either produces alone.
The market without the government produces maximum value extraction for those with market power and a progressive restriction of freedom for those without it. The government without the market produces an accumulation of red tape and an institution optimizing for its own perpetuation in the absence of a brutal feedback loop.
The mixed economy was always this argument. Not a compromise exactly, but a division of two complementary hypotheses which push on the levers the other structurally cannot.
The specific failure of the last few decades is not that the market dominated or that the government became ineffective.
It is that the government’s hypothesis became unclear, while the market’s hypothesis never wavered.
The Compass Without The Hypothesis
Having a compass is necessary, but it is not sufficient.
FDR had his famous formulation: “take a method and try it, if it fails admit it frankly and try another, but above all try something.” It is the most memorable description of experimental governance that I know. It is also, taken alone, potentially dangerous.
Bold persistent experimentation without a hypothesis is not experimentation, but iteration. And iteration without a hypothesis produces local optimization at best, producing the same coughing and fever experts we’ve already discussed. You get better at doing the thing you’re doing without ever asking whether you’re doing the right thing.
The scientific method’s specific genius is the hypothesis. The hypothesis does three things simultaneously that political argument almost never does. It states what you are testing for. It specifies in advance what would count as success. And, crucially, it specifies in advance what would count as failure. The hypothesis makes the experiment falsifiable before it runs. Which means the failure is information rather than catastrophe. You learned something about the territory rather than merely failing to impose your map.
FDR’s experimentation worked because the implicit hypothesis was consistent across experiments even when never formally stated. Does this improve the economic security and stability of ordinary people? The National Recovery Administration failed that test and was abandoned. Social Security passed it and was kept and built upon.
It is the difference between a direction and a test. The compass says something like “we are heading toward maximizing freedom for the most people.” The hypothesis says something like, “We predict that this specific policy, through this specific mechanism, will expand freedom for these specific people, and we will know if it has succeeded or failed by these specific measures.”
Without the hypothesis the compass produces sincerity without accountability. The policy was intended to expand freedom. It can always be argued that it did, somewhere, somehow, for someone. The failure is never definitive because the success was never specified.
With the hypothesis, the compass produces what democratic governance is actually supposed to produce: a record, a test, or a result that can then be honestly assessed.
Trump: FDR Without A Hypothesis
The most instructive contemporary demonstration of what bold persistent experimentation looks like when it has no hypothesis—and arguably no consistent compass—is the Trump administration.
The action is genuine. No serious observer denies that things happen, decisions get made, executive orders get signed, and operations get executed swiftly.
The Venezuela operation. The Iran strikes. The tariff announcements. The DOGE cuts. The speed and decisiveness are not performance. They are genuine expressions of a governing style that prioritizes action over deliberation.
What is absent is the hypothesis.
Is the goal of tariffs trade-imbalances, a re-establishment of domestic manufacturing, or revenue-generation? What level of inflation would count as a failure? By what measure, and what timeframe, would the administration acknowledge their failure or success?
Without the hypothesis, the tariffs become merely an episode generating its own momentum, defending themselves against contrary evidence because there are no hypotheses being tested.
But this is not a uniquely Trumpian problem. It is just the extremely visible form of a governing pathology that has run across several administrations and across parties, in many countries. The Trump version is simply legible in a novel way that more polished administrations conceal behind process and Orwellian double-speak.
Which is why FDR was of a different character. He tried the National Recovery Administration. It failed by the test the New Deal had implicitly committed to. It did not expand the practical freedom and economic security of ordinary people. He admitted it frankly, discarded it, and tried another.
That is bold persistent experimentation with a compass and a hypothesis. It is the governing method the current moment is demanding, but not quite receiving.
When Governments Optimize For The Wrong Thing
More insidious than governing without a hypothesis is optimizing for the wrong thing.
The American healthcare system is the canonical example. It is genuinely extraordinary at specific, measurable things that aren’t necessarily healthcare. Billing codes are processed with remarkable efficiency. Insurance claims are adjudicated with precision. Administrative costs are tracked and reported. The system has optimized itself extensively against the hypothesis it has actually been running: does this maximize the financial sustainability of its own institutions?
What is missing is its ostensible hypothesis: are Americans receiving the best medical care, for the most amount of people, for the least amount of money? It fails this repeatedly.
By some estimates, 45,000 - 65,000 Americans die annually from lack of health insurance. Medical debt is the leading cause of personal bankruptcy. Life expectancy in the United States trails other wealthy countries that spend less per capita.
Canadian housing is the same story at a different scale. The planning and approval systems that govern housing construction in Vancouver, Toronto, and increasingly Edmonton and Calgary are impressive administrative achievements for the wrong thing. Environmental reviews are thorough. Community consultation processes are comprehensive. Heritage assessments are detailed. Appeal mechanisms are accessible. The system is very good at what it has optimized for, which is preventing specific harms to specific people.
It is bad at what many would think “housing policy” should do: build houses. The average approval process in Canada takes a year—which is down, admittedly, by about 2 months—with Toronto taking 25+ months. Compare that to Tokyo, the most populated city in the world, which remains comparatively affordable, it takes weeks. The cost of carrying land and financing through that process adds substantially to the price of every unit eventually built. The consultation process that was designed to give communities a voice in their neighbourhood has become a mechanism through which incumbent property owners restrict the supply of housing to maintain their property values.
That is not to dismiss the value of these responses. They are almost aways a response to a genuine harm or concern. But the cumulative effects of the protections have produced a system that is running against the wrong hypothesis while the compass points in a direction the system cannot reach. The recovery is not to remove the protections, but to subordinate them to the primary goal.
Carney’s C-5 (One Canadian Economy Act) is an attempt at exactly this recovery. It does not eliminate the sub-hypotheses embedded in existing protections. It subordinates them to the primary one. Cabinet declares a project of national interest, and the question becomes not whether every individual protection has been satisfied but whether this project will serve the freedom and capability of most Canadians.
Whether it succeeds will be measured by whether housing gets built, infrastructure gets approved, and the total freedom of Canadians to live out the lives they value expands. Those are falsifiable tests. The record will exist. The both/neither on C-5 is that it is both a genuine attempt at hypothesis recovery and a political signal that will only be judged by its outcomes. That is what having a hypothesis means.
The Experiment
With the compass established and the hypothesis specified, FDR’s formulation recovers its full meaning.
Take a method, try it, then ask ourselves, “does this expand freedom for the most people?”
If it fails, admit it not as a catastrophe, but as information. Because the system was designed for imperfection. It has baked failure into the basis of all analysis. But we must try, anyway.
Then try another, adjusted and informed by the failure previously revealed, and update your hypothesis. The experiment runs again. The map improves incrementally towards the territory it is trying to map without ever claiming to have fully arrived with complete accuracy.
But above all, try something. Because an alternative future that contains only perpetual consultation, reworked frameworks, processes that are self-justified as necessary, cannot continue.
So, the try is not optional. The compass points toward a territory that requires navigation, then movement, then a willingness to be wrong, to remap and reorient.
This is what democratic governance is for. It is an ongoing experiment of pointing a compass and testing a hypothesis towards the maximization of freedom for the most amount of people.
The Both/Neither On Government
The honest conclusion is not that government or markets are the answer. It is that the government is the institution with the democratic mandate to hold the compass, state the hypothesis, and run the experiment on behalf of the people whose freedom it is supposed to maximize. Whether it performs that function well depends simply on the outcomes of that governance.
The right is correct that government can and does restrict freedom through overreach, through process-capture, through the institutional tendency to optimize for its own perpetuation rather than for the outcomes the compass requires. Every example in this essay is of policy that drifted from its ostensible hypothesis to a more measurable substitute, leading to a failure our first principle sets.
The left is correct that the alternative to government, which includes the unmanaged market, unregulated private power, and the absence of collective institutional capacity, does not automatically produce the freedom the compass points toward. The corporation that monopolises essential services or goods can restrict freedom as effectively as any regulatory burden. The labour market without the counter-balancing power of the government will inevitably restrict the worker’s capability. The housing market without supply, where the single measurable outcome is home-value, will inevitably restrict the freedom of individuals to own a home.
Both observations are correct. Both are pointing at real features of the territory. The both/neither on government is that it is simultaneously the institution most capable of expanding freedom for the most people and the institution most prone to veering from its stated mandate. The first principle, then, must be continually made clear, as a compass from which to direct policy, and as a standard against which to stop poor policy from metastasizing into the wrong hypothesis.
The recovery—always, repeatedly, as the drift recurs—is the same. Return to the compass.
The government’s role is to maximize freedom for the most amount of people.
Everything else is ongoing, never finished, and always requiring a renewal of effort and investigation at every level of complexity
