How Do You Measure the Performance of Politicians and Civil Servants?

I. The Unanswered Question

There is a question that ought to be asked in every democracy and is asked in none: how do you tell whether a politician is doing their job well?

In the private sector the answer, for all its qualifications, is clear: by results. The baker is measured by whether they bake good bread. The engineer by whether their bridge holds. The entrepreneur by whether their company survives. There is a product, a customer, an outcome. One can debate whether the outcome is the right measure — but there is a measure.

In politics there is no measure. There is no customer, no product, no market that feeds back whether the performance is adequate. There are only elections — every four years a mood that depends on a thousand factors, of which the actual performance of the elected official is one of the least significant. People vote for faces, parties, moods, fears. They vote against something more often than for something. And they vote without knowing what the elected official has concretely done in the past four years — because nobody measures it and nobody reports it.

The politician knows this. They know they are not measured by their performance but by their perception. And so they optimise not their performance but their perception. This is not a moral failing. It is the rational response to a system that knows no performance measurement.

II. The Civil Servant and Their Invisible Product

With the civil servant the problem is even sharper, because the civil servant is not even subject to the fiction of an election. They are appointed, not elected. They remain, regardless of their performance. They cannot be dismissed — not de facto, not in the reality of German civil service law, which has elevated irremovability to a constitutional principle.

The civil servant has a product, but this product is invisible: the functionality of the state. The road that is passable. The school that teaches. The court that delivers justice. The building permit that is granted. The pension that arrives on time. All of these are achievements of the civil service apparatus — but they become visible only when they are absent. When the road has a pothole. When the school cannot find a teacher. When the court takes three years. When the building permit takes eighteen months.

The civil servant lives in a system that knows their performance only ex negativo: through its absence. A well-functioning state is invisible. And whoever works invisibly is not measured.

III. The Surrogate Metrics

Where no real measure exists, surrogate metrics emerge. And the surrogate metrics of the public service are a study in institutional self-reference.

The civil servant is measured by their years of service — not by what they achieved in those years. They are measured by their career grade — not by the quality of their decisions. They are measured by the number of cases they processed — not by the quality of the processing. They are measured by their evaluation by their superior — who is evaluated by the same criteria and who has the same incentive to remain inconspicuous.

Evaluation in the public service is a masterwork of tautology: a civil servant assessed by a civil servant, who is assessed by a civil servant, according to criteria devised by civil servants, in a system administered by civil servants. There is no point at which an external reality — a citizen, a result, a measurable improvement — penetrates this loop.

The politician has their surrogate metric in media resonance. Not whether they solved a problem is measured, but whether they spoke about the problem. Not whether the law works, but whether the press conference went well. Not whether the infrastructure functions, but whether the ground-breaking ceremony was photographed. The politician's performance becomes a performance — in the theatrical sense of the word.

IV. The Proper Measure

If one honestly asks by what the performance of politicians and civil servants should be measured, there is only one answer that does the subject justice: by the functionality of the polity for which they are responsible.

Does the state function? Do the schools, courts, hospitals, roads, agencies, police, fire services, water supply, energy supply, telecommunications function? Can citizens conduct their affairs in a reasonable time and with reasonable effort? Can businesses be founded, invest, produce without being suffocated by bureaucracy? Can inventors realise their inventions? Can children grow up in safety and receive an education worthy of the name?

This is not an idealistic programme. It is the minimum. It is the core task of the state and therefore the core task of every individual in its service. And if one applies this measure — the plain functionality of the polity — then it becomes apparent that the performance of many politicians and civil servants in Germany has not been satisfactory for years.

Building permits take months. The digitalisation of public administration is a quarter-century behind. Schools are losing staff and quality. Bridges are decaying. Trains do not run on time. Broadband expansion stagnates. Asylum procedures take years. Courts are overloaded. The armed forces are not operational. The healthcare system groans. The pension system faces a demographic abyss.

This is not a polemical list. These are measurable facts. And for every single one of them, someone is responsible — a politician who bears the responsibility and a civil servant who is responsible for implementation. But neither is ever measured by these results. The politician is measured by the next election. The civil servant is measured by their next evaluation. And both have learned that it is safer to manage failure than to eliminate it.

V. The Self-Reference of Power

The deeper problem is that those who should be measured themselves determine how they are measured. The Bundestag passes the laws by which members of parliament are paid, provided for, and evaluated. The ministries determine the criteria by which their civil servants are promoted. The state governments decide how the schools they run are assessed. The state examines itself — and generally finds itself satisfactory.

This self-reference is the opposite of performance measurement. Performance measurement requires an external yardstick — a customer, a market, an independent auditor. In the private sector, the market is this auditor: whoever delivers poorly loses customers. In the state there is no market. The citizen cannot switch their state — not in the way they can switch their baker. They are the compulsory customer of a monopolist that grades itself.

Franz Oppenheimer described this structure over a century ago: the state as the organisation of the political means — an institution that lives from the productive labour of others and determines the conditions of its own existence. What Oppenheimer formulated as historical analysis describes the current situation with alarming precision.

VI. The Incentive System

People respond to incentives. This applies to entrepreneurs, engineers, and bakers — and it applies to politicians and civil servants. If one wants to understand why the public service functions as it does, one must examine the incentives under which it operates.

The civil servant is not rewarded for solving a problem. They are rewarded for not making a mistake. This is a fundamental difference. Problem-solving requires risk, initiative, deviation from the rulebook. Faultlessness requires the opposite: strict rule compliance, securing cover, obtaining as many signatures as possible before any decision is taken. The best civil servant — in the system's terms — is the one who attracts the least attention. Who produces no complaints, provokes no queries, takes no decisions that might be challenged.

The politician is not rewarded for solving a long-term problem. They are rewarded for looking good in the short term. The legislative period of four years is too short for most real problems — infrastructure, education, demographics, the energy transition — and too long for the media attention span. So the politician concentrates on what becomes visible within four years: announcing laws rather than implementing them. Promising money rather than changing structures. Naming problems rather than solving them. And above all: attributing blame for grievances to the previous government, which in turn did the same with its predecessor.

This incentive system produces with mathematical reliability precisely the outcome we observe: active, visible busyness alongside actual standstill. Many press conferences, few results. Many laws, little effect. Many civil servants, not much state.

VII. The True Performers of the State

They exist, the genuine high performers in public service. The teacher who, despite a shortage of colleagues and a surfeit of bureaucracy, teaches her pupils something. The judge who, despite overload, delivers careful judgments. The case officer who processes the building application in four weeks instead of eighteen months because she takes her job seriously. The police officer who stands on the street at night so that others may sleep.

These people exist. They are even common. And the system punishes them — rather than rewarding them. For whoever works faster than average receives not more pay but more work. Whoever judges better than their colleagues is not promoted but burdened. Whoever shows initiative risks making their superior feel bypassed. The system levels downwards, because excellence in public service is a reproach to mediocrity — and mediocrity is in the majority.

The self-proclaimed high performers, by contrast — the heads of department, the directors, the permanent secretaries, the ministers — are not measured by the functionality of what they lead but by their ability to make the dysfunction invisible. Their job is not to keep the state functional. Their job is to convey the impression that the state functions. And as long as this impression is maintained, the high performer is satisfied — and so is the system.

VIII. What Could Be Measured

It is not as though performance in the public sector were immeasurable. There are countries that do it.

Singapore pays its ministers on a performance basis — linked to GDP growth and the median wage. One can debate the criteria, but not the principle: there is a measurable outcome, and compensation depends on it.

Estonia has digitalised its administration — not as a press-conference project, but as a functioning state. Ninety-nine per cent of public services are available online. Processing times are measurable and partly published.

Switzerland has direct democracy — not a perfect system, but one that regularly and on concrete questions brings citizens to the ballot box, creating a considerably finer feedback loop than a general election every four years in which everything is decided at once and therefore nothing is really decided.

One could, if one wished, measure in Germany too. How long does a building permit take on average, and how has the duration changed under the responsibility of this minister, this department head, this section leader? How many pupils achieve age-appropriate reading competence after four years of primary school, and what has the education minister concretely done to improve this number? How many kilometres of motorway are repaired per year, and how does that relate to the budget allocated?

All of this is measurable. All of this is publishable. All of this could feed into a performance assessment worthy of the name. But nobody does it — because those who would have to do it are those who would be measured by it.

IX. The Circle

This is the circle that makes the problem appear insoluble: the reform of performance measurement would have to be decided by those who profit most from the absence of performance measurement. The Bundestag would have to pass a law that measures members of parliament by results. The ministries would have to issue regulations that evaluate civil servants by outcomes. The parties would have to nominate candidates willing to be measured against concrete targets.

None of this will happen. Not because those involved are wicked, but because they act rationally. No member of parliament votes for a system that might endanger their re-election. No minister issues a regulation that exposes their own record. No civil servant devises an evaluation system that would make them look bad.

Oppenheimer would have recognised this circle for what it is: the political means in its purest form. A class that lives from the labour of others and shapes the rules so that this way of life may endure. The only difference from the feudal structures Oppenheimer analysed is that the modern civil-service state no longer rules with swords but with forms — and that it does not call its rule governance but "public welfare provision".

X. The Only Authority

If the system will not reform itself — and it will not reform itself — there remains only one authority that can force performance measurement: the citizen.

Not the citizen as voter, who ticks a party box every four years and then watches for four years. But the citizen as principal, who asks concretely, measurably, and publicly: what have you done? What is the result? How has the functionality of the polity changed under your responsibility — for the better or for the worse? And who does not accept that the answer consists of excuses, jurisdictional deflections, and references to the difficult fiscal situation.

This sounds naive. It is naive. But the alternative is what we have: a state whose functionality nobody measures, whose responsible officials evaluate themselves, and whose citizens bear the result — in the literal sense. They are the true high performers. They bear the burden of an apparatus that does not know its own performance, because it has never measured it.

The question of how to measure the performance of politicians and civil servants is therefore not an academic question. It is the most political of all questions. For whoever determines the measure determines the power. And as long as there is no measure, power is uncontrolled.

A state without performance measurement is like a company without bookkeeping: it does not know where it stands, it does not know where it is going, and it only notices it is bankrupt when the money runs out.

For companies, there is then the insolvency administrator. For states, there is then history.