The In-Between Realm

I. A Missing Word

There is no good word for what sits between the funds and their purpose. Bureaucracy misses it, because an authority has a mandate you can look up. Administration misses it, because what is administered is a stock, and what happens here is passing something on. Apparatus sounds like intent, and intent is rarely at play.

We propose another word: Zwischenreich, an in-between realm. It names a layer that stands between the funds and the purpose, that lives by passing things through, and whose share is measured by throughput rather than by result. An in-between realm is nothing malicious. It does mediate; without it, in many cases nothing would arrive at all. That is what makes it so hard to attack.

It has, however, a property one has to know: it grows with the funds and does not shrink with them. When funds rise, throughput rises with them, and so does the need to mediate. When funds fall, the mediation stays the same — applications do not get simpler, verifications do not get shorter, panels do not get smaller. An in-between realm is elastic in one direction and rigid in the other.

II. How the Funding Landscape Grew

Not by design. By sedimentation.

Every crisis and every political goal of the past seven decades has left a programme behind: post-war reconstruction, the coal crisis, structural change, reunification, the Lisbon strategy, the financial crisis, the energy transition, digitalisation, Covid, the Zeitenwende. Hardly any of them was ever ended. What ended was usually the occasion, not the instrument; the instrument was given a new justification.

So the funding landscape today lies open like a geological section. The federal funding database covers programmes of the German federation, of the Länder, and of the European Union; the European Commission's own central overview lists the programmes financed from the Multiannual Financial Framework 2021–2027 and from NextGenerationEU by heading and cluster. Anyone applying today digs through sediments whose reasons for existing sometimes go back decades.

III. What Actually Happens There

Application. Peer review. Approval. Drawdown. Interim report. Proof of use. Final report. Audit of proof of use.

Read that sequence again and ask what it is about. It is about the use of the funds, throughout. And this is not arbitrary; it follows from the law. Section 44 (1) of the Federal Budget Code (BHO) ties every grant to an explicit stipulation: „Dabei ist zu bestimmen, wie die zweckentsprechende Verwendung der Zuwendungen nachzuweisen ist.“ — it must be laid down how the correct use of the grant is to be demonstrated. Proof of use is the statutory obligation. Whether anything came into being is not required with the same strictness.

The grantor is liable for the correct use of public funds; there are rules, audit procedures, and case law for that. No one is liable for the effect, because there is no yardstick for effect that would hold up in court. So what can be checked is what is checked — and what can be checked becomes the object.

From this follows a shift no one has decided on: the purpose „something is meant to come into being“ quietly becomes the purpose „it is to be accounted for correctly“. The two are not exclusive. But only the second has consequences.

IV. Who Defines What Happens

Here lies the core.

Programme lines are set in the ministries. Execution lies with Projektträger — project-management agencies which, on behalf of the federation, receive applications, have them reviewed, approve them, and settle the accounts. The legal construction for this stands in § 44 (4) BHO: „Juristischen Personen des privaten Rechts kann mit ihrem Einverständnis die Befugnis verliehen werden, Verwaltungsaufgaben auf dem Gebiet der Zuwendungen im eigenen Namen und in den Handlungsformen des öffentlichen Rechts wahrzunehmen &…“ — legal persons of private law may be granted the authority to carry out administrative tasks in the field of grants in their own name and in the forms of public law. Out of this enfeoffment a business has grown. The largest agencies — among them the Projektträger Jülich at Forschungszentrum Jülich, the DLR Projektträger, VDI/VDE Innovation + Technik GmbH, the AiF Projekt GmbH — together move billions. Their remuneration hangs on throughput, not on outcome.

Peer review is done by panels drawn from universities, institutes, and companies. And here a circle closes that is unavoidable in small fields: applicants and reviewers are one and the same layer of people. Anyone who knows a field well enough to judge an application in it will, as a rule, be filing applications in it himself. This is not corruption but a numbers problem — there are not enough experts to keep the two roles cleanly separate. The result is nevertheless a selection that confirms what the layer already holds to be right.

We have met the same figure elsewhere: in the standardisation of an invention methodology, the same persons write the standard, hold the certification levels, and sell the courses. Nothing there is unlawful either. And there too the result is that a rival school finds no way in.

V. Who Controls

The Courts of Audit. And their work shows exactly where the boundary runs.

In a 2024 report the Federal Court of Auditors (Bundesrechnungshof) revisited the success monitoring of funding programmes, and the sunsetting and degression of subsidies under the government's Subsidy Policy Guidelines. The report's title names the procedure for what it is: „Part 2: Audit of Success Monitoring of Funding Programmes – Follow-Up Audit“. A follow-up audit means: there was a first audit, it produced findings, and the second look was to check whether they had been remedied. That such a report was needed years after the first finding says enough about the inertia of the process.

The Court itself audits predominantly the propriety of expenditure — again, the use of funds. Economy audits and success audits are the exception, even though § 7 (2) BHO requires „appropriate economy studies“ for all financially significant measures. The duty is on the books. Whether it is honoured is another matter.

Then there are evaluations. Most of them are drawn up by institutes that themselves live from the funding landscape and that, in turn, receive the contract to evaluate a programme. That is a second closed circle: the evaluated party finances the evaluator, and the evaluator needs the next commission.

And above all this stands a circumstance no one says out loud: a ministry that would end a programme without result would thereby also strike the apparatus running it and the institutes living off it. In this arrangement, no one's interest lies in establishing that a programme has come to its end.

VI. Why „Efficiency“ Is the Wrong Word

In this debate people speak endlessly of efficiency, and the word is at this point not merely imprecise but misleading.

Efficiency measures the ratio of input to output. But the output here is the report, the proof of use, the approved application. A programme can produce its reports highly efficiently and be completely ineffective. To increase efficiency, in such a case, is to speed up precisely what has no effect anyway.

The concept that matters is effectiveness: has a capability arrived in the stock? Is it available to a user who did not develop it herself?

For this there is a simple and very uncomfortable test:

What would not have happened if the programme had not existed?

This question can be answered. It requires no econometric study, only the readiness to accept, in some cases, nothing as the answer — that a project was funded which would have gone ahead anyway, or one that never arrived anywhere even with funding. It is therefore rarely asked.

Three further tests, no more expensive:

Who carries it on? A result with no bearer stays put. This question is neither asked in the application nor answered in the final report.

What is still missing now? Between a research result and its application there regularly sits something small and unfinanced: a standard, a proof, a reference plant. Such gaps appear in no programme line because they are too small to justify a programme — and too important to let pass.

What has been ended? A system without closure is not a system but a sediment. The count of ended programmes would be the most telling figure of the whole landscape. It is not kept.

VII. What the Realm Systematically Filters Out

A selection procedure that examines applications does not pick out the best projects but those best suited to the procedure. These are not the same.

Application-ready is whoever masters the language, knows the deadlines, has prior work to show, an institution at his back, and the pre-financing in place. Application-ready is, above all, whoever can make a statement about the market.

And with that, the procedure reliably filters out precisely the class of projects for which it was set up. For something truly new the market cannot be quantified because it does not yet exist: nothing is burning as long as the solution is missing. Whoever makes the market question a condition of entry systematically selects what confirms the existing order.

The free inventor is not disadvantaged in this procedure. He simply does not appear in it. There is no programme line whose object is a person.

VIII. The Objection We Take Seriously

Everything above can be turned around, and we do it here ourselves.

Without this layer no public money would arrive anywhere. The use audit prevents misuse, and it does so effectively; cases of embezzled grant funds are rare, and that is an achievement. The panels prevent funds from being handed out by political whim. The forms compel someone to at least think a project through once.

And: much has in fact come into being. Whole research fields, institutes, training programmes, infrastructures. To deny this is to make it too easy on oneself.

The objection strikes the existence of the layer — not its property of being measured by throughput. And it does not answer the question from section VI: what would not have happened? As long as it is not asked, the defence of the layer is as unsubstantiated as its indictment.

IX. What Follows

Not an abolition programme. Four proposals that all cost little.

First: a sunset clause. Every programme ends on a fixed date and must, for any extension, be justified anew — not with reports about the use of funds but with an answer to the question of what would not have happened without it. The Subsidy Policy Guidelines of the German government already require sunsetting and degression of subsidies; the Federal Court's follow-up audit shows that these requirements do not bite in practice. They do not bite because they lack a hard endpoint.

Second: a bearer question in the application. A field stating who will carry the result on, and what is meant to move that bearer. If it remains blank, the project is not fundable — or it is run as pure basic research, which would be more honest.

Third: responsibility for the last mile. For standards, proofs, and reference installations that are too small for a programme and too important to be left out. A fund from which such things can be paid without an application procedure.

Fourth: separated evaluators. Evaluations may not be produced by institutions filing applications in the same programme area. This is self-evident for financial auditors, and it is not here.

X. Conclusion

The in-between realm is not the work of bureaucrats. It is the sum of reasonable individual decisions, each of them right in itself: that public funds should be audited, that experts should review, that misuse must be prevented, that every political goal needs an instrument.

What is missing is the opposite force. There is no one whose task it would be to establish that something is over. That is why the layer grows in one direction and does not shrink in the other.

And that is why, at the end of a funding landscape in which agencies, ministries, and institutes keep one another alive, the result is that a finished capability fails to enter the stock for forty years because a standard sheet is missing that would have made no one rich.