n.V2-6.01| WHO CAN ACTUALLY SAY YES |

01| Imagine a group of residents in a barrio who have decided to rebuild the place themselves. They are not asking to be helped; they have a plan, they have people willing to do the work, and what they need from government is permission, a small amount of money, and for nobody to obstruct them. They go to the municipal offices, because that is the government they can physically reach, and they explain what they intend to do. The official they meet may be sympathetic, may even be enthusiastic. And at the end of the meeting he will very likely tell them that he agrees with all of it and cannot approve any of it.

02| That answer is not evasion, and understanding why it isn't is the beginning of understanding government. He genuinely cannot approve it, because the money that would pay for it is not his money and the rules that govern its use were written somewhere else. To see why, you have to stop treating "the government" as a single thing. It is at least two things, they sit in different buildings, they answer to different electorates, and they hold entirely different powers. Nearly everything that looks like incompetence or indifference from the barrio is actually the friction between those two.

03| The first is the national government — the ministries, the treasury, the parliament. It writes the laws, raises most of the taxes, and decides what the money may be spent on. The second is what the literature calls subnational government: the province or region, and below it the municipality, the city, the district. This is the tier that runs the clinic, collects the waste, issues the permit, and knows the name of the street. When a poor person meets the state, it is nearly always this tier they are meeting.

04| Now the shape of the problem. The tier that can see the barrio has very little money, and almost none of it is its own. Across 135 countries, subnational governments together account for "over 21.5% of total public spending around the world" [OECD/UCLG 2022, p. ??] — every province, city and district on earth, spending about a fifth of the public purse between them. And most of what they spend was not raised locally: subnational governments "still depend largely on grants for more than half of their revenue" [ibid.]. A grant is money sent down from the national government. It usually arrives earmarked — meaning tagged for a named purpose, and unusable for anything else.

05| So the municipal official in paragraph 01 is not being difficult. He may be running a budget in which nine tenths of the money arrived pre-committed to purposes decided in the capital, none of which is "a group of residents with a plan." The one thing he possesses that the capital does not — knowing exactly which street, which family, which problem — is the one thing the funding system has no format to receive. And the capital, for its part, is not ignoring the barrio: it is designing against a national average, using statistics that arrive months late from the same municipalities that depend on it for money. It has statistical offices, not doorbells.

06| This is not a local quirk. It has a name in the research — unfunded mandates, meaning responsibilities handed down without the money to discharge them. Rodríguez-Pose and Vidal-Bover measured the gap across 518 regions in 30 OECD countries between 1997 and 2018, by scoring how much political authority each region held and how much fiscal capacity, and subtracting one from the other. They found that "the presence and dimension of unfunded mandates undermines economic growth" [Rodríguez-Pose & Vidal-Bover 2024, p. ??], and — this is the important part — the damage comes from the mismatch itself, not from being centralised or decentralised. A region can be highly devolved and function; what hurts is being told to do things you have not been funded to do.

07| They also found that the gap is widest where places are poorest: unfunded mandates are "more prevalent in regions with relatively lower levels of development" [ibid.]. Which produces the ugly arithmetic at the centre of this chapter. The poorer the barrio, the poorer the municipality it sits in; the poorer the municipality, the more dependent it is on earmarked grants; the more dependent it is, the less discretion it has to say yes to anything that was not already decided in the capital. Need and authority run in opposite directions.

08| For residents with a plan, the practical consequence is that there are two signatures on any yes, not one. There is the tier that wants it — the municipality that can see why it makes sense — and there is the tier that controls the money it would be paid from. Those are different offices, they are usually not talking to each other about this, and the second one has never seen the barrio. Any strategy for approaching government that treats the mayor as the decision-maker will stall, not because the mayor is unwilling, but because he was never the whole of the decision.

09| The obvious response is that governments should simply push money and power downward, and some have. This is where it gets genuinely difficult, because two countries did exactly that and got opposite results. Denmark is the strong case: municipalities and regions there handle 65% of all public expenditure, and municipalities set their own income tax rate, which supplies 88.7% of their tax revenue. The tier that sees also decides, and it decides with money it raised itself — though even there, 58% of local resources still arrive as transfers, so the gap is narrowed rather than abolished.

10| Bolivia tried the same move in a much poorer setting. The 1994 Law of Popular Participation doubled the municipal share of national tax revenue from 10% to 20%, distributed strictly per head, and created 315 municipalities where there had been 186. Money did reach neglected peripheries. But the same reform "exacerbated the problem of clientelism" in rural areas, fragmented Guarani collective decision-making, and set communities competing against one another for the funds. Money moved down; power did not follow it in the way the law assumed, and in some places what arrived instead was a new supply of favours for whoever controlled the municipal office.

11| The dilemma is therefore real and unresolved, which is why governments oscillate rather than settle. Centralise, and the decisions are made far from the people affected by them. Decentralise, and the decisions are made close to the people — including close to whoever locally is strongest. There is no arrangement that removes the trade; there are only different placements on it. What Denmark had and Bolivia did not was not more money but functioning local accountability to sit underneath it, and that is not something a transfer formula can send.

12| So this is the first of the forces, and it is not a verdict on government. Read it as a variable that differs from city to city. It is strong — the gap wide, the municipality powerless — where local own-source revenue is a small share of the budget, where transfers arrive earmarked, and where the municipality cannot set any tax rate itself. It is weak where a local tax is genuinely levied and collected, where transfers arrive as blocks rather than tagged lines, and where the municipality has some discretionary budget of its own. Before entering anywhere, that is what to look at first, because it determines who has to be persuaded and whether the person in front of you can say yes at all.