Uphill money: why the EU budget makes poorer households pay for richer ones

Why this is on your desk now#

The EU is negotiating its budget for 2028 to 2034, and the summit on 15 October 2026 broke up without a deal. Two days earlier a German government official told Reuters that Germany pays “a little over 40 billion” now and would still be “over 60 billion” a year under the compromise the Irish Council presidency had just put forward. On 16 October FOCUS readers let rip about it: Europe as a “shunting yard for German money”.

Two details matter before anyone gets angry at the wrong number. The 60 billion is no published calculation; it comes from an anonymous official. And it doesn’t refer to the Commission’s original proposal but to the Irish compromise, which is already 8% smaller. If the smaller version still costs Germany 60 billion, think about what the bigger one costs.

The chancellor’s position is clear, at least in words. “We are still far from an agreement,” Friedrich Merz said on 10 October. Two days later: “Not having a new financial framework at all is by far the cheapest solution for Germany.” Leaders meet again on 26 and 27 November.

So here are three things nobody in Brussels likes to put on one page: where the money comes from, where it leaks away, and why the key that decides who pays what is unfair and outdated. Then a better one. You can set the weights yourself.

Thirty-five years of “about 1%”#

The EU plans its money in seven-year frameworks. For decades the size barely moved as a share of the economy:

The table hides the real change. Since 2020 the EU borrows. NextGenerationEU, the corona recovery fund, was €806.9 billion in today’s money, financed with joint bonds. EU debt went from €52 billion in 2019 to €738.9 billion at the end of 2025 and €836.8 billion in the middle of 2026. The European Court of Auditors expects about €1 trillion by the end of 2027.

That debt is now hitting the budget. Interest for 2021 to 2027 was planned at €14.9 billion. The auditors now expect about twice that. In 2025 the interest line was €4.96 billion, in 2026 it’s €7.97 billion and the draft for 2027 says €9.9 billion. Repayment starts in 2028 and runs until 2058. €421 billion of the grants paid out have to come back out of future budgets. Germany’s implicit share of that is about €103 billion. It received about €30 billion of grants itself.

Where the money comes from#

The EU has four sources of its “own” money. Three are small, one does the heavy lifting.

1. Customs duties. Collected at the border. Member states keep 25% as “collection costs”, the rest goes to Brussels. That’s about 11% of the budget.
2. A VAT-based resource. 0.3% of a harmonised VAT base, capped at half of national income. About 13%.
3. The plastics levy. €0.80 per kilo of plastic packaging that isn’t recycled. About 4%. Seventeen poorer countries get fixed discounts.
4. The GNI-based resource. Every country pays the same percentage of its gross national income. This is the balancing item: whatever the other three don’t cover.

Guess first

What share of the 2026 EU budget is financed through that GNI-based “membership fee”?

Show me

How the membership fee is set, with the numbers of the 2026 budget:

The rebates are the political part. Denmark, Germany, the Netherlands, Austria and Sweden get fixed reductions, Germany €3,671 million a year in 2020 prices. All 27 pay for them by GNI share, the rebate countries included. So Germany’s gross rebate of €4.56 billion in 2026 shrinks to a net €2.25 billion.

Note the wording of the rebate clause in the law: “for the period 2021-2027”. If no new revenue decision is ratified, the old one keeps running, but in my reading without the rebates. Merz’s “no deal is cheapest” may be less clear-cut than it sounds. I’m not a lawyer, and nobody in Berlin has published a calculation.

Who pays what#

The Commission stopped publishing net balances per country after 2019. Inconvenient numbers tend to disappear that way. Using its own published method on its own data for 2025:

Germany’s net balance on this method was −€14.9 billion, 0.32% of its national income. Counting its share of the recovery grants that have to be repaid, it was −€24.5 billion. Other methods give anything from €13 to €25 billion, which is why every politician can quote the number that suits them.

What leaves the German federal budget in cash is easier to read. EU payments came to €32.0 billion in 2024, €34.9 billion in 2025, an estimated €44.3 billion in 2026 and €46.1 billion in the 2027 draft. The finance ministry itself blames the fall in federal tax revenue in 2026 “mainly” on higher payments to the EU.

The next seven years: two trillion and new taxes#

The Commission’s proposal of 16 July 2025 asks for €1,984.9 billion for 2028 to 2034, 1.26% of EU income. €168 billion of that is debt repayment.

What else is in it:

• New EU taxes worth €58.5 billion a year in 2025 prices: a share of CO2 trading, the carbon border levy, a levy on electronic waste, a cut of tobacco excise and a new charge on companies with more than €100 million turnover. The Court of Auditors notes that only €13.6 billion of that does not come out of national budgets anyway. National contributions would still rise by 81% to €235 billion a year.
• A higher ceiling. What the EU may call from member states rises from 1.40% to 1.75% of national income, with extra headroom for debt and crises: up to 2.60% in all.
• All rebates gone, and member states keep only 10% of customs instead of 25%.
• More borrowing, on top of the ceilings: up to €395 billion of crisis loans and €150 billion for a new “Catalyst Europe” loan fund.

The Irish compromise of 10 October trims the total by 8% to €1,826 billion. It’s still 30% above the current framework. Germany’s quarter of €235 billion in national contributions alone is about €58 billion, before customs and the new levies. That’s where the 60 billion comes from.

Where it leaks#

Now for the part the Commission prefers not to have on the same page. The best witness against EU spending is the EU’s own Court of Auditors.

• Adverse opinion seven years running. For every financial year from 2019 to 2025 the auditors found the spending so riddled with errors that they refused to confirm it was legal and regular. Spending has never received a clean opinion since the Court started this audit in 1994.
• 3.8% paid out against the rules in 2025, about €5.0 billion. In 2023 it was 5.6%, about €9.0 billion. Regional funds 6.6%, agriculture 3.9%, rising.
• The Commission stopped publishing its own overall error estimate in 2024, in exactly the years the auditors found farm errors going up. If the thermometer shows fever, take the thermometer away.
• The recovery fund pays estimates, not costs. In 15 of 19 measures checked the work cost less than estimated, and the country got the full estimate anyway. France got €500 million for a measure that cost €368 million. 9 of 37 payments in 2025 broke the payment conditions; 10 milestones were paid without being met. The auditors’ verdict: “not a performance-based instrument”. Climate spending in it was overstated by €34.5 billion.
• Fraud. The EU prosecutor had 3,602 open cases at the end of 2025 with €67.3 billion of estimated damage. Most of that is VAT fraud that hits national treasuries. 512 cases concern the recovery fund, €5.1 billion. Over ten years, criminal courts recovered about €180 million.
• Administration. €13.28 billion in 2026, up 14% since 2024. The Commission wants €118 billion for 2028 to 2034, up from €82 billion. About 60,100 staff in the institutions plus 16,100 in 38 agencies. Frontex alone went from €134 million in 2015 to €1.2 billion. The EU’s pension liability for its staff is €91.1 billion, and there is no pension fund behind it.
• Farm money. About 20% of recipients get 80% of direct payments.
• Red tape. The Commission itself puts the recurring paperwork cost of its rules at €150 billion a year for businesses and wants to cut €37.5 billion of it. That cost doesn’t show up in any budget, but you pay it in prices and wages.

How much is that per year? Errors are not theft, and interest is not fraud, so I add up only what clearly should not have been paid or did not have to be: money paid out against the rules, customs fraud, avoidable administration such as the second parliament seat in Strasbourg, unearned recovery fund payments and the interest on the joint debt.

This is an estimate with stated assumptions, not an audit figure. The middle case is more than the EU spends on its entire administration. And it leaves out projects that follow every rule and are still useless. When the auditors checked 20 EU-funded airports, 9 of them hadn’t needed the money at all.

Worse than the money: the rulebook#

Money that leaks is gone once. A rule costs every year, in every firm, for as long as it stands. And it kills the things you never see: the company that’s never founded, the hire that’s never made, the factory that goes to Texas instead. That’s why I think Brussels’ rulebook does more damage than all the money it wastes.

• €150 billion a year. That’s the Commission’s own rough estimate of the recurring paperwork cost of rules for businesses in the EU, and it counts only the paperwork, not the lost output. It’s about ten times the middle of my waste estimate above. The Commission’s goal is to cut a quarter of it, €37.5 billion. So far it claims €18 billion of savings, mostly in proposals that aren’t law yet.
• 13,000 acts in five years. Mario Draghi counted about 13,000 EU legal acts from 2019 to 2024, against about 3,500 federal laws and 2,000 resolutions in the US. Only 515 of the 13,000 went through the normal law-making process. The rest are implementing and delegated acts that no parliament voted on one by one.
• “One in, one out” barely works. The Commission’s own net savings from that rule fell from €7.3 billion in 2022 to €0.26 billion in 2024.
• Firms say it themselves. 55% of small and medium-sized firms call regulation and administrative burden their biggest challenge. More than 60% of companies see regulation as an obstacle to investing.
• Per firm: data protection compliance costs up to €500,000 for a small company and up to €10 million for a large one, sustainability reporting €150,000 to €1 million.
• Germany adds its own layer. Germany’s regulatory control council reported this October that the cost of complying with rules rose by €1.3 billion, mainly because of how Germany transposed the EU’s sustainability reporting rules. Three EU-driven laws passed in 2025 add about €5 billion of compliance cost on top, and they aren’t even in that count yet. For EU implementing acts, as a rule, nobody calculates the cost at all.

The best proof that much of it was never needed: in 2026 the EU cut the sustainability reporting duty for about 80 to 90% of the firms it had applied to. Two years after the rules came in. The cheapest way to cut red tape is not to write it.

Bill Gates walks into the room#

Now to the key itself. Imagine ten people in a room, each with €100,000. Bill Gates walks in. The average wealth in the room jumps to over €10 billion. On average, everyone is a billionaire. Nobody in the room is any richer.

That’s roughly how the EU measures who can pay. GNI per head is an average. It counts the profits of large companies, the salaries at the top and the income of the state’s own enterprises. It says nothing about what the household in the middle has.

Look at the household in the middle and the picture turns around:

Guess first

The median household in Malta owns how many times as much as the median household in Germany?

Show me

Germany’s median household (€103,300) owns exactly as much as the Estonian and practically as much as the Greek one. The median household in Slovakia, Croatia, Czechia, Spain, Italy and Belgium owns more. Malta, Cyprus, Slovenia, Czechia, Croatia, Portugal and Spain are all net recipients or close to zero; their typical households are richer than Germany’s.

To be fair, there are reasons for this. German wealth is concentrated: the top 10% hold 54 to 59% of it, in Malta about 35%. 58% of German households rent, and the median renter household has €18,400. State pension claims aren’t counted anywhere. And on average, a German adult still owns about twice as much as a Maltese one. But that’s exactly the point: the average is Bill Gates. The question is what the typical citizen who pays the taxes owns.

Four more things the GNI key gets wrong:

• It reads Ireland’s tax-haven bookkeeping as wealth. Ireland’s GNI is €437.7 billion. Without the paper profits of redomiciled multinationals, the Irish statistics office’s own GNI* is €334.0 billion. The key overstates Ireland’s capacity by about 31%.
• It is regressive at the edges. The rebate countries pay 0.56 to 0.66% of their income. Hungary and Lithuania pay 0.84%. The plastics levy costs Hungary 0.10% of its income and Germany 0.03%.
• It doesn’t see public debt. Italy’s median household owns 58% more than Germany’s, its state owes 137% of GDP. Belgium: median household 2.5 times Germany’s, state debt 108%. Spain: 1.5 times, 101%. The GNI key treats rich citizens of indebted states the same as poor citizens of solid ones. And when the debt goes wrong, the rest of the euro area carries the risk.
• It ignores the future. By 2050 Greece will have 68 people aged 65 and over for every 100 of working age, Italy 61, Germany 46, Ireland 43. A country that will soon have far fewer workers per pensioner can pay less tomorrow than its GNI suggests today.

A flow measure from the 1980s, applied to a union that now has debt, a shared currency and very different demographics. Outdated is the polite word.

Not just Germany: all 27 on the hill#

Germany is the biggest case, not the only one. Line up all 27 members by what their typical household owns and look at who paid and who got money in 2025. The yardstick is the EU’s typical household: about €136,800, the average of the national medians weighted by the number of households. Richer than that and paying, or poorer and receiving: the money flows downhill, as it should. Everything else flows uphill.

In 2025 five countries whose typical household is richer than the EU’s received €4.7 billion net. Three countries whose typical household is poorer, Germany, Austria and Finland, paid €16.1 billion net. A few cases stand out:

• Luxembourg and Belgium look like the worst cases and aren’t. Luxembourg received €2,854 million from the EU budget in 2025. €1,957 million of that is the EU’s own administration based there, which the net balance already leaves out. Of the remaining €896 million, only about €172 million went through the usual national channels: farms, regions, the Brexit reserve. The rest are EU programmes paid to bodies and companies that happen to sit in Luxembourg, from the IT behind the customs and tax systems to satellite connectivity and EU agencies. The statistics book that money where the recipient sits. Counted only by what reaches the country itself, Luxembourg pays in more than twice what it gets. Belgium is the same story with Brussels. So both stay out of the uphill count, which makes the rest of the table more solid, not less.
• Malta and Cyprus: households worth €376,400 and €240,600 in the middle, and €364 and €278 per inhabitant from the budget.
• Finland and Austria are on Germany’s side of the hill. Finland’s typical household owns €96,000, less than the Greek one at €103,200. Finland paid €70 per inhabitant, Greece received €370.
• Italy and Spain look fair here: rich households, net payers into the regular budget. But they’re also states with debt above 100% of GDP, and once their recovery fund grants are counted, both were net recipients in 2025: Italy by €2.0 billion, Spain by €2.8 billion. The rest of the EU starts repaying those grants in 2028.
• France pays, as its households’ wealth says it should. Its risk sits elsewhere: 116% public debt in a shared currency. That’s what the debt premium in my key is for.
• Ireland pays, too, but on an income inflated by multinational paper profits. Its contribution is 0.68% of GNI, but 0.89% of the income its economy actually earns (GNI*). Germany pays 0.63%.
• Poland, the biggest recipient in euros at €6.7 billion, isn’t in the ECB survey. My estimate puts its typical household at about Germany’s level. If that holds, it’s another uphill case. Until there’s real data, I leave it on the fair side.
• Where it works: Latvia, Lithuania, Bulgaria, Romania, Hungary, Greece, Croatia, Slovakia and Estonia. Poorer households, net money. That’s what a common budget is supposed to do, and nobody should object to it.

A fairer key#

My proposal, in one sentence: split the bill by what people earn, what the typical household owns and how well people actually live, then correct for debt and ageing.

Why each piece is there:

• Income stays in, because contributions are paid in euros from current income. But at 40%, not 100%.
• Typical wealth is the Bill Gates correction: the median, not the average. A household that owns a paid-off flat and has no rent to pay can carry more than a renter with the same income. I don’t weight it more than 40%, because a house doesn’t pay cash and the median favours countries of owners over countries of renters.
• Living standard measures what people actually consume per head, with local prices taken into account. That’s where currency and price levels come in: a euro buys more in Sofia than in Copenhagen.
• The debt premium applies the EU’s own treaty limit of 60%. Whoever borrows above it puts the shared currency and the joint debt at risk and pays for that risk. Italy and Belgium pay more, Germany a little, Estonia nothing. No reward for running up debt.
• The ageing discount is small on purpose. Ageing reduces future capacity, but it’s also the result of decades of policy. 25% of full weight seems right to me. Set your own.

Inflation doesn’t get a separate weight, because it’s already in the nominal income figures and in the price levels behind the living standard. Competitiveness, infrastructure and education are needs, not capacity: they belong on the spending side, see below.

Play with it. Pick “Income only” to see roughly today’s logic without rebates, pick “Typical wealth only” for the pure Bill Gates correction, or set your own weights:

Build a fairer EU key: who would pay what

Income only (today’s logic)My keyTypical wealth only

1. Which yardstick decides who pays how much? the three shares always add up to 100%

Share split by income

%

By what the country earns in a year (gross national income). That’s how it works today. Ireland counts without the paper profits of multinationals (GNI*).

Share split by typical wealth

%

By what the household in the middle owns: home, savings and investments minus debts. Middle, not average, so a few billionaires don’t make a country look rich.

Share split by living standard

%

By how well people actually live: what they consume per head, adjusted for local prices.

2. Corrections on top

Surcharge per 10 points of debt above 60%

%

The treaty limit is 60% of GDP. At 2%, a country with debt at 100% of GDP pays 8% more. Whoever borrows above the limit puts the shared currency and the joint debt at risk.

Discount for ageing

%

Countries that will have more pensioners per worker in 2050 than the EU average pay less, the others more. 0 = ignore ageing, 100 = full weight.

Your country🇦🇹 Austria🇧🇪 Belgium🇧🇬 Bulgaria🇭🇷 Croatia🇨🇾 Cyprus🇨🇿 Czechia🇩🇰 Denmark🇪🇪 Estonia🇫🇮 Finland🇫🇷 France🇩🇪 Germany🇬🇷 Greece🇭🇺 Hungary🇮🇪 Ireland🇮🇹 Italy🇱🇻 Latvia🇱🇹 Lithuania🇱🇺 Luxembourg🇲🇹 Malta🇳🇱 Netherlands🇵🇱 Poland🇵🇹 Portugal🇷🇴 Romania🇸🇰 Slovakia🇸🇮 Slovenia🇪🇸 Spain🇸🇪 Sweden

Country	Change per inhabitant and year	Pays today (2025)	With your key	Net balance: today → your key
🇱🇺 Luxembourg	+€350	€375m	€613m	+€550m → +€312m
🇲🇹 Malta	+€187	€169m	€276m	+€209m → +€102m
🇸🇪 Sweden	+€88	€3.5bn	€4.4bn	−€1.3bn → −€2.2bn
🇧🇪 Belgium	+€79	€4.6bn	€5.6bn	+€622m → −€315m
🇨🇾 Cyprus	+€65	€266m	€330m	+€273m → +€209m
🇮🇹 Italy	+€21	€17.4bn	€18.6bn	−€3.3bn → −€4.5bn
🇩🇰 Denmark	+€19	€2.8bn	€2.9bn	−€483m → −€596m
🇨🇿 Czechia	+€17	€2.4bn	€2.6bn	+€2.4bn → +€2.2bn
🇬🇷 Greece	+€11	€1.9bn	€2.0bn	+€3.8bn → +€3.7bn
🇷🇴 Romania	+€9	€2.9bn	€3.1bn	+€5.9bn → +€5.8bn
🇸🇮 Slovenia	+€7	€511m	€526m	+€349m → +€334m
🇫🇷 France	+€5	€23.0bn	€23.3bn	−€5.4bn → −€5.7bn
🇵🇹 Portugal	+€4	€2.4bn	€2.5bn	+€1.5bn → +€1.4bn
🇳🇱 Netherlands	+€4	€6.7bn	€6.8bn	−€3.2bn → −€3.3bn
🇧🇬 Bulgaria	+€4	€894m	€918m	+€1.4bn → +€1.3bn
🇵🇱 Poland	+€2	€7.0bn	€7.0bn	+€6.7bn → +€6.6bn
🇸🇰 Slovakia	−€1	€989m	€983m	+€1.3bn → +€1.3bn
🇪🇪 Estonia	−€3	€313m	€309m	+€573m → +€577m
🇦🇹 Austria	−€9	€3.1bn	€3.1bn	−€851m → −€772m
🇭🇷 Croatia	−€9	€757m	€721m	+€1.1bn → +€1.1bn
🇪🇸 Spain	−€11	€13.4bn	€12.9bn	−€1.1bn → −€543m
🇭🇺 Hungary	−€20	€1.8bn	€1.6bn	+€2.1bn → +€2.3bn
🇩🇪 Germany	−€30	€29.0bn	€26.5bn	−€14.9bn → −€12.4bn
🇱🇻 Latvia	−€35	€328m	€262m	+€1.1bn → +€1.2bn
🇱🇹 Lithuania	−€38	€689m	€578m	+€1.4bn → +€1.5bn
🇫🇮 Finland	−€91	€2.2bn	€1.7bn	−€394m → +€121m
🇮🇪 Ireland	−€111	€3.0bn	€2.4bn	−€512m → +€93m

The pot stays the same: the €132bn of national contributions in 2025. Customs duties stay where they are. Net balance = what flows back minus what is paid in (Commission data 2025, own calculation, administration not counted). Median wealth: ECB household survey 2023; Bulgaria, Denmark, Poland, Romania and Sweden are not in the survey and are estimated. Infrastructure: share of firms calling transport a major obstacle to investment (EIB survey 2025). “Uphill” marks the countries on the other side of a net transfer from your country that runs from poorer to richer on both counts. Belgium and Luxembourg are left out of that check, because EU bodies and contractors based there inflate their net figures. A model, not a law.

With my weights the pot stays at €132.4 billion, but it lands differently. Luxembourg pays €350 more per inhabitant a year, Malta €187, Sweden €88, Belgium €79, Cyprus €65. Germany pays €26.5 billion instead of €29.0 billion, €30 less per inhabitant. With “typical wealth only” Germany would pay €20.5 billion and Malta, Luxembourg and Belgium would become net payers.

My key has weak spots too, and you’ll find them in the table. Finland and Ireland move to around zero with my weights, Finland mainly because of a low median wealth. Malta would still get slightly more back than it pays in. That’s why the key alone isn’t enough. The other half of the budget is what the money is for and where it goes. The calculator already checks one rule from that half, the uphill check: more on that below.

What the money should be for#

A budget without a goal grows until it runs out of other people’s money. So first the goals, then the rules.

Goal 1: competitive, not kept alive. Subsidies that never end don’t rescue an industry, they freeze it. Farming is the prime example. Direct payments and market support take €284 billion of the current framework, 23% of everything, and about 20% of recipients get 80% of the direct payments. That’s not a policy for farmers, it’s an annuity for owning land. My rule: every support programme gets an end date. Transition money for one seven-year period at most, a seventh less every year, then zero. It pays for what makes a business viable without help: modernisation, productivity, new markets. Where an investment earns money, the EU lends or takes a stake instead of giving it away, so the money comes back. If an industry can’t stand on its own after seven years of help, a second seven years won’t change that. And the cheapest help costs nothing: open markets, and the €37.5 billion of red tape the Commission itself wants to cut.

Goal 2: lift the bottom, wherever it lives. The target isn’t countries with a low average, it’s households that have little: the typical household, measured by median wealth and living standard, region by region. That includes the renter in Duisburg or Gera, not only the smallholder in Bulgaria. The tools are conditions, not cash: transport, broadband, schools and more housing, so rents stop eating the income.

Goal 3: infrastructure that connects everyone. Rail, power grids, pipelines and data links across borders. A single market needs them, and no country builds them alone. Ranked by benefit per euro, not by national quota.

Who pays for whose roads?#

The rule of thumb in one line: nobody pays uphill. A country whose typical household is poorer and whose infrastructure is worse should never be a net payer for a country that is better off on both counts. And everyone uses their own money first.

To compare infrastructure across all 27, the best measure I found is the EU investment bank’s company survey: the share of firms that call transport infrastructure a major obstacle to investing. It’s a survey of complaints, not an engineering report, but it’s the same question everywhere. In Germany 25% of firms say so, in Czechia 3%, in Belgium 6%.

Now run Germany through the test with the 2025 numbers. Germany paid €14.9 billion net. In the same year the budget paid net money to five countries whose typical household owns more than a German one and whose firms complain less about roads, rail and bridges:

Together that’s €6.6 billion a year. Germany pays about 22% of all national contributions, so roughly €1.5 billion of German money flows uphill every year. Belgium and Luxembourg would be on this list too, but their net figures are inflated by EU bodies and contractors based there, so I leave them out. The calculator above runs the same check for whichever country you pick.

How to fix it, in three rules:

1. No uphill transfers. If a net payer is worse off on both counts than a net recipient, the recipient’s envelope shrinks until that flow is gone. Malta and Cyprus don’t trip this rule, because their firms complain even more about transport than German ones. The next rule catches them anyway.
2. Own money first. The EU pays at most the share by which a country’s typical household is poorer than the EU’s typical household, about €136,800 (the average of the national medians, weighted by households). Never more than 80%. The rest comes from the country itself. Public debt is no reason to get money: a state that has already borrowed beyond 60% has to cut elsewhere, not send the bill to Brussels.

France, with debt at 116% of GDP, gets nothing for its roads: its citizens own more than the EU’s typical household, and its state’s debts are its own business. Germany, by the same rule, could get up to a quarter of the cost of its crumbling bridges. That’s what lifting the bottom wherever it lives means in practice.

1. Sunset and payback. All support for a sector ends after seven years, falling every year. Money that earns a return is lent, not given.

And three rules for the budget itself:

1. No payment without proof. Payment only against actual costs, not estimates. If the error rate in a programme is above 2%, payments stop until it’s fixed. That’s the auditors’ own materiality threshold.
2. Debt first. No new EU borrowing before the €421 billion is on a credible repayment path. Any new debt needs a vote in every national parliament, not just in Brussels.
3. A ceiling that means it. 1% of national income, like for thirty years, including interest and repayment.

When the job is done, or can’t be done#

Every goal above has an end, and that’s the point. If the goals are reached, the typical household in Latvia or Bulgaria close to the EU’s, the cross-border networks built, the industries standing on their own, then the transfers have done their job. If they’re not reached after two budget periods, fourteen years, the programmes have failed. Either way the budget shrinks afterwards, it doesn’t grow. Success is no reason to keep paying, and failure is no reason to pay more.

What remains is the core that the EU was founded for and does better than its members: the single market and its rules of the game, competition law, trade policy, the customs union and the external border. Everything else goes back to the countries, where voters can see what it costs and vote it out. My guess is that this core costs a fraction of today’s 1% of national income, and needs a fraction of today’s rulebook.

This isn’t a plan to leave the EU. It’s a plan to make the EU pay for what it promised when it was sold to us: a single market, open borders, a stable currency. Not a transfer union that gets bigger every seven years and leaks a little more each time.

What this has to do with you#

If you live in Germany, the 60 billion isn’t an abstract number. Someone pays the difference between €35 billion and €60 billion, and the federal budget has no €25 billion lying around. It can come from three places: higher taxes, more debt or cuts elsewhere. My bet is on all three, starting with the “temporary” solidarity surcharges and levies that never go away.

What I do with that:

• Plan with a higher tax burden, not a lower one. Every promise of relief comes with a “subject to financing”.
• Don’t rely on the state for anything you can arrange yourself, especially your pension. The EU’s budget planners now have to finance €9.9 billion of interest a year. Your pension fund has the same problem.
• Spread your bets across currencies and countries. The euro area is a community of liability, whether that’s in the treaty or not.

What if I’m wrong?#

Maybe the single market brings Germany so much that €60 billion is a bargain. Some FOCUS readers think so. But the benefit of the market doesn’t depend on the budget: Norway and Switzerland have access without paying a quarter of the EU’s bills, and the EU didn’t need two trillion euros to create the single market. Maybe the rebates do survive, and maybe the Irish compromise is cut further. Then Germany pays 50 billion instead of 60. That’s still a key that charges the renter in Gelsenkirchen and pays out to the homeowner in Valletta.

This article is education, not investment, tax or legal advice. It contains my opinion, an estimate with its assumptions and a model with the formula shown.

Sources#

• FOCUS online, “Europa wird zum Verschiebebahnhof für deutsches Geld”, 16 October 2026
• Reuters via onvista, “Berlin warnt bei EU-Finanzrahmen: Zahlungen könnten auf mehr als 60 Milliarden steigen”, 13 October 2026
• Federal Government, statement by Chancellor Merz on the Irish presidency’s proposal, 10 October 2026
• Euronews, “Zoff um EU-Geld”, 15 October 2026 (Merz: no framework is cheapest)
• Irish Presidency, proposal for the EU’s long-term budget, 10 October 2026
• Council Decision (EU, Euratom) 2020/2053 on the system of own resources
• EU budget 2026, definitive adoption (revenue tables, GNI rate)
• European Commission, EU spending and revenue, data 2000–2025 (29 September 2026)
• Commission, technical adjustment of the MFF for 2027, COM(2026) 299
• European Parliament fact sheet, multiannual financial framework
• Commission proposal for the MFF 2028–2034, COM(2025) 571
• Commission proposal for a new own-resources decision, COM(2025) 574
• European Court of Auditors, Opinion 04/2026 on the own-resources proposal
• European Court of Auditors, overview of the MFF 2028–2034 proposals, April 2026
• European Court of Auditors, annual report on the financial year 2025, 8 October 2026
• European Court of Auditors, 2025 audit in brief
• European Court of Auditors, review 02/2025 on the recovery fund’s performance model
• European Court of Auditors, special report 21/2014 on EU-funded airport infrastructure
• European Court of Auditors, special report 14/2024 on climate spending in the recovery fund
• Commission, draft budget 2027, statement of estimates (interest, administration)
• Commission, half-yearly report on borrowing, H1 2026
• EPPO annual report 2025
• OLAF report 2025, impact of investigations
• Commission, working documents to the draft budget 2027 (staff, agencies)
• EU annual accounts 2025 (pension liability)
• EIB Investment Survey 2025, transport infrastructure as an obstacle to investment
• EPRS briefing on CAP direct payments, July 2026791452.html)
• Commission, simplifying EU rules, 24 June 2026 (administrative burden)
• Commission, communication on simplification and implementation, COM(2025) 47 (€150bn administrative cost, €37.5bn target)
• Mario Draghi, The future of European competitiveness, part B, September 2024
• Nationaler Normenkontrollrat, annual report 2026
• Federal Ministry of Finance, monthly report August 2026, overview of the current situation
• Bundesbank, monthly report October 2025, EU finances 2024
• IW Köln, EU budget and member states: net payers and net recipients, November 2025
• ECB, Household Finance and Consumption Survey, wave 2023, statistical tables (October 2026)
• ECB, Distributional Wealth Accounts
• UBS Global Wealth Report 2026
• Eurostat, actual individual consumption per head in PPS
• Eurostat, government debt
• Eurostat, population projections EUROPOP2023 (old-age dependency 2050)
• Eurostat, number of households
• CSO Ireland, annual national accounts 2025 (GNI*)

Education, not advice. I don’t know your situation, and past returns promise nothing. Check my numbers, then make your own call. You’re a grown-up.