Question 3 of 8

How will the state pay for defence and care when EU funds shrink?

About 20 min read

The short answer

If nothing changes, the state does not go bankrupt, but it does not choose either. Defence at five per cent of GDP, Rail Baltica and dearer borrowing keep the deficit above three per cent of GDP for most of the 2030s. The European Union's rules then force savings: about 2.8 points of GDP by 2040, spread over the budget by a formula, not by a decision. Government debt reaches about 61 % of GDP in 2040 and falls afterwards. Where exactly it peaks, between 59 and 69 %, depends on how the labour market behaves in the 2030s.

The fall in EU funds after 2028 can be covered from any tax base. Each brings in more than the fall takes: labour taxes about 11 points of GDP over 2028–2045, nearly three times the loss. Under the rules that buys fewer forced cuts in the 2030s, not a lower debt in 2040. The choice between the bases is a choice about who pays. Labour taxes fall on earners and feed envelope wages. Consumption taxes fall on the poorest. Capital and property taxes fall on owners and on investment.

Why it matters

Three things are already decided, and together they set the next ten years of the budget. First, the starting point. The year 2025 closed with a recorded deficit of 2.5 % of gross domestic product (GDP, that is, everything the country produces in a year). The 2026 budget plans 3.3 %, and the European Commission expects 4.3 % for 2027. The model carries that plan: its deficit reaches 4.0 % of GDP at the start of 2028. Second, three committed items. Defence spending at five per cent of GDP on the NATO definition. The state's share of Rail Baltica, about 0.6 % of GDP a year from 2026 to 2034, 4 points of GDP in all. And interest, which rises from 1.1 % of GDP now to 1.9 % in 2035 and 2.5 % in 2040, as the era of cheap borrowing ends. Third, the rules. Latvia has adopted the European Union's fiscal framework. Once the structural deficit, that is, the deficit with the ups and downs of the economy taken out, stays above three per cent of GDP, the framework's corrective arm requires savings every year until it is back below.

On the base path that happens from 2028, and the rule keeps working through the 2030s. The reason is a slump the model's own labour market sets off. Retirements open posts faster than the shrinking workforce can fill them, so by 2030 unemployment is low and wages rise. The economy then cools: output runs below its potential in the mid-2030s, and the deficit stays above three per cent, 3.4 % of GDP in 2035 and 3.8 % in 2038. Public spending, 48 % of GDP in 2030, is 46 % in 2040: that fall is the rule at work, not a policy anyone has chosen. Debt rises from 47 % of GDP at the end of 2025 to about 62 % in 2039 and is 61 % in 2040.

The base path assumes the shadow economy, that is, economic activity hidden from the tax authorities, stays where it is, at about 22 % of GDP. Bringing it onto the books is a policy, and it is on the old-age income page as the formalisation package. Revenue therefore stays near 44 % of GDP throughout, and everything that moves is on the spending side.

Government debt

What the state owes, as a share of what the country produces in a year.

Recorded data
Base path
Uncertainty band

The grey band shows how far the base path moves when the two most important assumptions are varied across their plausible ranges.

Show the dataDownload CSV

Year	Recorded data	Base path	Uncertainty band – low	Uncertainty band – high
2005	12.4	15.0
2006	10.6	11.3
2007	8.9	8.1
2008	19.5	10.4
2009	38.0	24.9
2010	48.6	47.9
2011	46.6	48.4
2012	44.0	43.2
2013	41.8	42.0
2014	43.1	42.2
2015	38.3	40.9
2016	41.7	39.9
2017	40.3	40.0
2018	38.3	38.4
2019	37.9	36.8
2020	44.0	39.2
2021	45.9	41.9
2022	44.4	43.5
2023	44.4	42.3
2024	46.2	43.5
2025	46.9	44.5	44.5	44.5
2026		45.4	45.3	45.4
2027		47.7	46.5	48.1
2028		49.8	48.2	50.5
2029		50.9	47.4	51.7
2030		52.0	48.1	53.2
2031		53.4	47.9	55.3
2032		54.1	48.8	56.8
2033		54.9	50.2	57.5
2034		55.8	51.6	57.9
2035		57.2	54.7	58.0
2036		58.7	57.6	58.7
2037		60.3	58.7	61.6
2038		61.1	58.7	64.6
2039		61.7	58.9	67.5
2040		61.2	58.9	69.0

On the chart the dots are the recorded data and the line is the model's base path. The grey band is wide: from 59 to 69 % of GDP in 2040. It shows how far debt moves when the two assumptions that matter most here are varied across their ranges. Both are about the labour market of the 2030s: how easily firms replace workers with equipment when workers get scarce, which sets how much wages rise and how deep the slump is, and how long firms take to open the posts they want. Neither has a Latvian record behind it. So the level of debt in 2040 is a point under a band, and the band is the honest answer; the ranking of the tax bases below holds across the whole of it. The model's figures are read at the start of each year: its 2025 debt is 44 % of GDP, the position at the start of 2025, against the record's 47 % at the end of the year. On the level of debt the fit is good; on the deficit and spending in the crisis years it is weaker, see "What to keep in mind".

Why it matters beyond the number. Debt is not a goal in itself. The goal is a state that can pay for defence, doctors and schools when the funds run out, and that chooses what to cut or tax rather than having a rule choose for it. The question this page answers is therefore not "how much to save" but "which tax base carries it, and who pays".

What drives it

Public finances have two feedbacks and one rule, and on this base path the rule is the strongest of the three.

The debt snowball: a reinforcing circle. More debt means larger interest payments, larger interest payments a larger deficit, and a larger deficit more debt. It is a circle that strengthens itself, and it is exactly how countries end up in debt crises. In Latvia it has been weak, because interest rates were low and debt moderate. That is ending. The interest bill rises from 1.1 % of GDP in 2025 to 2.5 % in 2040 as old cheap bonds are replaced by new ones at about four per cent. The circle also runs the other way: once debt starts shrinking, interest shrinks with it, which is why the line comes down after 2040.

The debt feedback. Plus means "the more of one, the more of the other". All three links run the same way, so the circle strengthens itself: debt breeds debt, but shrinkage also breeds shrinkage. At four per cent interest it is no longer slow.

The tax base that gives way: a balancing circle. When taxes rise, part of the base disappears. Higher labour taxes mean more envelope wages, a higher value added tax more undeclared transactions, a higher profit tax less investment. That is why no tax collects as much as the arithmetic says, and why revenue is larger in the first years after a tax rise than ten years later. The labour tax package yields 1.4 points of GDP more revenue in 2031, but only 0.6 more in 2040.

The rule. The European Union's fiscal framework limits how fast public spending may grow. Once the structural deficit passes three per cent, it requires savings until the deficit is back below. On the base path the rule adopts 1.7 points of GDP of savings by 2035 and 2.8 by 2040. They fall on the functions that have no protection: general public services, support to the economy, education. It is not a feedback but a law, and it is the reason this page has no scenario called "borrow and wait". It is also why the packages below mostly change the size of that consolidation rather than the level of debt. A package that brings the budget money lets the rule cut less. By 2040 the labour tax package carries 2.5 points of GDP of rule-imposed savings against 2.8 on the base path.

The slump of the 2030s. This is the base path's own feature, and it rests on three judgements. How easily firms replace workers with equipment when workers get scarce: the less easily, the more wages rise and the deeper the slump. How long firms take to open the posts they want. And what share of the posts that departures leave is advertised again. None of the three has a Latvian record behind it; the first two are the band on the chart.

Over the next fifteen years the rule and the retreat of the tax base dominate. The debt snowball matters because interest is no longer cheap, and that is the strongest reason not to let debt rise further than the base path already does.

What can be done

Four policies with which parliament can cover the fall in EU funds, each with the same target: about one and a half per cent of GDP a year in extra revenue or savings, on paper. For each, the page says what it delivers in 2040 against the base path, in which the fall in funds is not covered, what it brings or costs the budget, and how fast it acts. The budget figure is the package's own effect on the budget balance, added up over 2028–2045 in points of GDP, before the rule responds. Beside it is the difference in government debt in 2040, which is small and sometimes of the opposite sign, because under the rules extra revenue mostly replaces cuts rather than paying down debt. All four bring the budget money. The price is paid by people and by the economy, and that is in the table under "Assessment".

Labour taxes

What it is
personal income tax a tenth higher, social contributions a sixth higher, and the contribution ceiling, above which no more is paid on a wage, twice as high.

What it delivers
10.9 points of GDP for the budget over 2028–2045, the most of the three bases. The at-risk-of-poverty rate, that is, the share of people whose income is below 60 % of the median, is 0.3 points lower in 2040, and inequality is lower. The most progressive of the three. Debt in 2040 is nevertheless +0.6 points against the base path, because the rule cut 10 points of GDP less over the years.

What it costs
the economy per head in 2040 is 0.8 % smaller. The shadow economy is 1.3 points of GDP larger. The pensioner poverty risk is 0.8 points higher, because envelope wages mean smaller pension contributions.

How fast it acts
in revenue at once, most in 2031, and part of it leaks away afterwards.
Consumption taxes

What it is
the standard rate of value added tax from 21 to 23 %, the scope of the reduced rates halved, excise duties a sixth higher.

What it delivers
11.0 points of GDP for the budget over 2028–2045. Debt in 2040 +0.6 points, for the same reason as above.

What it costs
the economy per head 0.7 % smaller. The at-risk-of-poverty rate 0.1 points higher and inequality higher, because the poorest spend a larger share of their income. The shadow economy 0.8 points larger. The only one of the three that raises poverty.

How fast it acts
in revenue at once, most in 2031.
Capital and property taxes

What it is
a 15 % tax on retained company profits, which are currently untaxed; cadastral values, frozen since 2016, revalued to market level, and the property tax rate one and a half times higher.

What it delivers
6.6 points of GDP for the budget over 2028–2045, less than the other two bases but durable: its revenue keeps growing after the rule has done its work. Inequality 0.3 points lower, the at-risk-of-poverty rate 0.1 points lower. The shadow economy unchanged. Collects nearly everything the arithmetic promises. Debt in 2040 +0.7 points.

What it costs
the economy per head in 2040 1.1 % smaller, the most of the three, because the profit tax reduces investment. Owners and investing firms pay.

How fast it acts
in revenue at once and durably; visible in the economy from 2038.
Spending reallocation

What it is
general public services, support to the economy and the remaining functions a tenth smaller, social protection other than pensions three per cent smaller.

What it delivers
19.8 points of GDP for the budget over 2028–2045, the most of the four, because no tax base gives way: there is none to give. Debt in 2040 −1.7 points: these are largely the same cuts the rule would have made anyway, taken earlier and by choice.

What it costs
the economy per head 0.1 % smaller in 2040 after a dip of 0.3 % in 2033. The at-risk-of-poverty rate almost unchanged. Those who receive the services pay, and the model does not see which ones.

How fast it acts
at once.

What the packages do

Five packages, all from 2028 with a three-year phase-in. In all five, EU funds from the 2028 budget period on are a third smaller than on the base path, which itself already lets them taper after 2034. In the first nobody covers it; three cover it from one tax base each; the fifth, a balanced policy, from all three bases a third each. Two more packages, spending reallocation and a revenue-neutral tax shift, have been calculated and are listed in the table below the charts.

Funds fall, unfinanced. EU funds a third smaller from 2028, and nobody covers it.
Labour taxes. the fall in funds, covered by income tax and social contributions.
Consumption taxes. the fall in funds, covered by value added tax and excise.
Capital and property. the fall in funds, covered by a profit tax and the property tax.
Balanced policy. the fall in funds, covered by all three bases a third each.

Debt. The hump is the same in every package, because what makes it is in the base path: the deficit of the late 2020s, the slump of the 2030s and the rule's answer to them. The unfinanced fall in funds leaves debt in 2040 at 63 % of GDP against 61 % on the base path. The tax packages lower debt a little in the early 2030s, by 0.2 points in 2035 with labour taxes, and then sit just above the base path: 61.8 % with labour taxes, 61.8 % with consumption taxes, 61.9 % with capital and property taxes, 61.7 % with the balanced policy. The lines cross because the base path keeps cutting for longer. A package that brings in revenue in 2028–2035 makes the rule adopt fewer savings, and those savings, once adopted, stay in force.

Government debt

What the state owes, as a share of what the country produces in a year.

Recorded data
Base path
Funds fall, unfinanced
Labour taxes
Consumption taxes
Capital and property
Balanced policy

Show the dataDownload CSV

Year	Recorded data	Base path	Funds fall, unfinanced	Labour taxes	Consumption taxes	Capital and property	Balanced policy
2015	38.3	40.9
2016	41.7	39.9
2017	40.3	40.0
2018	38.3	38.4
2019	37.9	36.8
2020	44.0	39.2
2021	45.9	41.9
2022	44.4	43.5
2023	44.4	42.3
2024	46.2	43.5
2025	46.9	44.5	44.5	44.5	44.5	44.5	44.5
2026		45.4	45.4	45.4	45.4	45.4	45.4
2027		47.7	47.7	47.7	47.7	47.7	47.7
2028		49.8	49.8	49.8	49.8	49.8	49.8
2029		50.9	50.9	50.6	50.7	50.7	50.7
2030		52.0	52.1	51.1	51.3	51.5	51.3
2031		53.4	53.5	51.7	52.0	52.6	52.0
2032		54.1	54.5	52.1	52.5	53.3	52.5
2033		54.9	55.6	53.2	53.5	54.4	53.5
2034		55.8	56.7	54.9	55.0	55.7	55.1
2035		57.2	58.3	57.0	57.0	57.6	57.1
2036		58.7	59.9	59.0	58.9	59.3	59.0
2037		60.3	61.6	60.8	60.7	61.0	60.7
2038		61.1	62.5	61.6	61.6	61.8	61.6
2039		61.7	63.1	62.2	62.1	62.3	62.1
2040		61.2	62.7	61.8	61.8	61.9	61.7

The deficit and revenue. The tax packages' effect on the deficit is largest in 2031, when they are at full strength and the base has not yet given way: labour taxes then cut the deficit by 1.0 points of GDP. From the mid-2030s the packages run a slightly larger deficit than the base path, because they carry fewer rule-imposed savings; by 2040 they are 0.1 points below it again. Revenue shows what the bases keep. The labour tax package yields 1.4 points more in 2031 and 0.6 in 2040; the capital and property package 0.8 in 2040 and still rising. Capital and property taxes keep nearly all of it, labour taxes less than half.

Government deficit

How much more the state spends than it collects in a year.

Recorded data
Base path
Funds fall, unfinanced
Labour taxes
Consumption taxes
Capital and property
Balanced policy

Show the dataDownload CSV

Year	Recorded data	Base path	Funds fall, unfinanced	Labour taxes	Consumption taxes	Capital and property	Balanced policy
2015	1.5	0.5
2016	0.0	0.2
2017	0.3	0.6
2018	1.4	0.7
2019	0.2	−0.7
2020	4.1	−0.7
2021	7.2	1.0
2022	4.9	3.4
2023	2.3	1.1
2024	1.8	0.8
2025	2.5	1.0	1.0	1.0	1.0	1.0	1.0
2026		2.6	2.6	2.6	2.6	2.6	2.6
2027		3.4	3.4	3.4	3.4	3.4	3.4
2028		4.0	4.0	4.0	4.0	4.0	4.0
2029		3.5	3.5	2.9	3.0	3.2	3.0
2030		3.4	3.4	2.5	2.7	3.0	2.7
2031		3.5	3.7	2.6	2.7	3.1	2.7
2032		2.7	3.1	2.5	2.6	2.8	2.6
2033		2.9	3.1	3.2	3.1	3.1	3.1
2034		3.2	3.4	4.0	3.8	3.7	3.9
2035		3.4	3.5	4.2	4.0	3.8	4.0
2036		3.7	3.8	4.2	4.1	3.9	4.1
2037		3.8	3.8	3.9	3.9	3.7	3.8
2038		3.8	3.8	3.5	3.6	3.5	3.5
2039		3.4	3.5	3.1	3.2	3.2	3.1
2040		2.6	2.6	2.4	2.4	2.4	2.4

Government revenue

Everything the state collects in taxes, contributions and other income, as a share of GDP.

Recorded data
Base path
Funds fall, unfinanced
Labour taxes
Consumption taxes
Capital and property
Balanced policy

Show the dataDownload CSV

Year	Recorded data	Base path	Funds fall, unfinanced	Labour taxes	Consumption taxes	Capital and property	Balanced policy
2015	38.5	39.7
2016	38.8	40.3
2017	39.3	40.4
2018	39.3	41.0
2019	39.5	42.0
2020	40.2	42.2
2021	39.3	42.0
2022	39.4	42.6
2023	41.1	42.4
2024	43.5	43.0
2025	43.6	43.4	43.4	43.4	43.4	43.4	43.4
2026		44.0	44.0	44.0	44.0	44.0	44.0
2027		44.2	44.2	44.2	44.2	44.2	44.2
2028		44.3	44.3	44.3	44.3	44.3	44.3
2029		44.9	44.9	45.5	45.4	45.2	45.4
2030		45.0	44.9	46.0	45.8	45.6	45.9
2031		44.8	44.7	46.2	46.0	45.6	46.0
2032		45.4	45.1	46.5	46.3	46.0	46.4
2033		45.3	45.0	46.2	46.1	45.9	46.1
2034		44.8	44.4	45.4	45.4	45.3	45.5
2035		44.4	44.0	44.9	44.9	44.9	45.0
2036		44.2	43.8	44.7	44.8	44.7	44.9
2037		44.2	43.9	44.7	44.9	44.8	44.9
2038		44.0	43.7	44.5	44.7	44.7	44.8
2039		43.7	43.4	44.3	44.5	44.5	44.6
2040		43.8	43.5	44.4	44.5	44.5	44.6

Interest. The interest bill is the quiet item on this page. On the base path it rises from 1.1 % of GDP in 2025 to 1.9 % in 2035 and 2.5 % in 2040. The packages change it by a tenth of a point either way. What changes it is the interest rate and the level of debt, and both are the base path.

Interest on government debt

What the state pays in interest on its debt in a year, as a share of GDP.

Base path
Funds fall, unfinanced
Labour taxes
Consumption taxes
Capital and property
Balanced policy

No comparable recorded series exists for this indicator; only the model's path is shown.

Show the dataDownload CSV

Year	Base path	Funds fall, unfinanced	Labour taxes	Consumption taxes	Capital and property	Balanced policy
2015	2.02
2016	1.75
2017	1.55
2018	1.33
2019	1.14
2020	1.06
2021	0.98
2022	0.94
2023	0.96
2024	1.06
2025	1.13	1.13	1.13	1.13	1.13	1.13
2026	1.20	1.20	1.20	1.20	1.20	1.20
2027	1.28	1.28	1.28	1.28	1.28	1.28
2028	1.37	1.37	1.37	1.37	1.37	1.37
2029	1.43	1.43	1.42	1.42	1.42	1.42
2030	1.50	1.50	1.47	1.48	1.48	1.48
2031	1.58	1.58	1.53	1.54	1.56	1.54
2032	1.65	1.66	1.58	1.59	1.62	1.60
2033	1.72	1.74	1.66	1.67	1.70	1.67
2034	1.81	1.84	1.77	1.78	1.81	1.78
2035	1.93	1.98	1.92	1.92	1.94	1.92
2036	2.07	2.13	2.08	2.08	2.10	2.08
2037	2.22	2.30	2.26	2.25	2.27	2.25
2038	2.35	2.45	2.40	2.40	2.41	2.39
2039	2.46	2.58	2.52	2.52	2.53	2.52
2040	2.53	2.67	2.61	2.61	2.61	2.60

The gain and the price for people. Lower debt is not a goal in itself. The goal is a state that can pay for what it has promised without lowering people's living standards. So here are two indicators that show what consolidation does to the economy and to people. The economy per head, measured in today's money, that is, at 2025 prices, grows on the base path from 23,100 euro in 2025 to 33,000 euro in 2040. Each tax package lowers it a little: labour taxes by 0.8 %, consumption taxes by 0.7 %, capital and property taxes by 1.1 %, because the profit tax reduces investment. These are small differences against the growth. But they are ordered the opposite way to what the budget keeps: the base that collects the most durably pays with growth.

GDP per head

The value of everything produced in Latvia in a year, per resident, in constant prices.

Recorded data
Base path
Funds fall, unfinanced
Labour taxes
Consumption taxes
Capital and property
Balanced policy

Show the dataDownload CSV

Year	Recorded data	Base path	Funds fall, unfinanced	Labour taxes	Consumption taxes	Capital and property	Balanced policy
2015	18,284	16,894
2016	18,914	17,738
2017	19,746	18,541
2018	20,765	19,390
2019	21,062	20,394
2020	20,462	20,923
2021	22,050	21,216
2022	22,680	22,027
2023	22,235	22,680
2024	22,424	23,114
2025	23,125	23,621	23,621	23,621	23,621	23,621	23,621
2026		23,909	23,909	23,909	23,909	23,909	23,909
2027		24,204	24,204	24,204	24,204	24,204	24,204
2028		24,805	24,805	24,805	24,805	24,805	24,805
2029		25,487	25,487	25,487	25,487	25,487	25,487
2030		26,147	26,147	26,138	26,140	26,142	26,140
2031		26,814	26,814	26,788	26,793	26,796	26,791
2032		27,409	27,406	27,363	27,371	27,372	27,366
2033		27,992	27,979	27,936	27,947	27,939	27,940
2034		28,528	28,486	28,486	28,490	28,469	28,484
2035		29,066	28,988	29,069	29,065	29,016	29,058
2036		29,634	29,516	29,649	29,639	29,568	29,628
2037		30,405	30,250	30,388	30,380	30,290	30,359
2038		31,260	31,068	31,146	31,154	31,055	31,117
2039		32,199	31,974	31,951	31,981	31,880	31,925
2040		33,028	32,870	32,763	32,809	32,671	32,735

The at-risk-of-poverty rate is almost flat on the base path, around 21 % in 2040. Labour taxes lower it by 0.3 points, because a higher income tax and a higher contribution ceiling fall more on the better off. Consumption taxes raise it by 0.1, because the poorest spend nearly everything they earn. Capital and property taxes lower it by 0.1. But labour taxes have a side effect this chart does not show: more envelope wages mean smaller pension contributions, and the pensioner poverty risk in 2040 is 0.8 points higher. The old-age income page explains why that matters.

At-risk-of-poverty rate

Share of people whose income is below 60 % of the median.

Recorded data
Base path
Funds fall, unfinanced
Labour taxes
Consumption taxes
Capital and property
Balanced policy

Show the dataDownload CSV

Year	Recorded data	Base path	Funds fall, unfinanced	Labour taxes	Consumption taxes	Capital and property	Balanced policy
2015	22.5	22.0
2016	21.8	21.9
2017	22.1	21.7
2018	23.3	21.8
2019	22.9	21.8
2020	21.6	21.8
2021	23.4	21.8
2022	22.5	21.8
2023	22.5	21.7
2024	21.6	21.6
2025	22.0	21.7	21.7	21.7	21.7	21.7	21.7
2026		21.8	21.8	21.8	21.8	21.8	21.8
2027		21.8	21.8	21.8	21.8	21.8	21.8
2028		21.7	21.7	21.7	21.7	21.7	21.7
2029		21.5	21.5	21.5	21.5	21.5	21.5
2030		21.4	21.4	21.4	21.4	21.4	21.4
2031		21.3	21.3	21.2	21.3	21.3	21.3
2032		21.3	21.3	21.1	21.4	21.2	21.2
2033		21.2	21.2	21.0	21.3	21.2	21.2
2034		21.2	21.2	20.9	21.3	21.1	21.0
2035		21.0	21.0	20.7	21.1	20.9	20.9
2036		20.9	20.9	20.6	21.0	20.8	20.7
2037		20.8	20.8	20.5	20.9	20.6	20.6
2038		20.7	20.7	20.4	20.8	20.6	20.6
2039		20.7	20.6	20.3	20.8	20.5	20.5
2040		20.6	20.6	20.3	20.8	20.5	20.5

Income inequality, measured by the Gini index, where zero means everyone has the same income and one hundred means one person has it all, says the same: labour and capital taxes lower it, consumption taxes raise it, the balanced policy changes it little.

Income inequality (Gini)

0 means everyone has the same income; 100 means one person has it all.

Recorded data
Base path
Funds fall, unfinanced
Labour taxes
Consumption taxes
Capital and property
Balanced policy

Show the dataDownload CSV

Year	Recorded data	Base path	Funds fall, unfinanced	Labour taxes	Consumption taxes	Capital and property	Balanced policy
2015	35.4	36.2
2016	34.5	36.0
2017	34.5	35.8
2018	35.6	35.8
2019	35.2	35.9
2020	34.5	35.8
2021	35.7	35.8
2022	34.3	35.8
2023	34.0	35.7
2024	34.2	35.5
2025	35.6	35.6	35.6	35.6	35.6	35.6	35.6
2026		35.7	35.7	35.7	35.7	35.7	35.7
2027		35.8	35.8	35.8	35.8	35.8	35.8
2028		35.6	35.6	35.6	35.6	35.6	35.6
2029		35.4	35.4	35.4	35.4	35.4	35.4
2030		35.2	35.2	35.2	35.3	35.2	35.2
2031		35.1	35.1	35.0	35.1	35.0	35.0
2032		35.0	35.0	34.8	35.2	34.9	34.9
2033		35.0	35.0	34.6	35.1	34.8	34.8
2034		34.9	34.9	34.4	35.0	34.7	34.7
2035		34.7	34.7	34.2	34.9	34.5	34.5
2036		34.5	34.5	34.0	34.7	34.3	34.3
2037		34.4	34.4	33.9	34.6	34.1	34.2
2038		34.3	34.3	33.8	34.5	34.0	34.1
2039		34.2	34.2	33.7	34.5	33.9	34.0
2040		34.2	34.2	33.7	34.5	33.9	34.0

All packages that were run

Package	Description	Government deficit 2040, points	Government debt 2040, points
no taper	The base path without the post-2034 taper of EU funds: receipts flat at the 2028-2034 level to 2050. A negative cost: what the taper on the base path alone adds to debt.	−0.1	−0.7
Spending reallocation	The fall in EU funds, covered by spending cuts: general public services, economic affairs and the remaining functions 10 % lower, social protection other than pensions 3 % lower.	+0.2	−1.7
Revenue-neutral tax shift	The fall in EU funds stays unfinanced; the tax-free allowance 30 % higher, paid for by value added tax at 23 % and a revalued cadastre with a property tax rate 1.3 times higher.	+0.1	+1.3

Assessment

Package	Budget effect 2028–2045, points of GDP	Government debt 2040, points	Revenue 2040, points of GDP	Economy per head 2040	Poverty rate 2040, points	Inequality 2040, points	Who gains, who pays	Confidence
Funds fall, unfinanced	−3.7	+1.5	−0.3	−0.5 %	0.0	0.0	nobody gains; the functions the rule cuts pay	high
Labour taxes	+10.9	+0.6	+0.6	−0.8 %	−0.3	−0.5	people in work pay, high earners the most; future pensioners and the declared economy lose	medium
Consumption taxes	+11.0	+0.6	+0.8	−0.7 %	+0.1	+0.2	everyone pays, the poorest relatively the most	medium
Capital and property taxes	+6.6	+0.7	+0.8	−1.1 %	−0.1	−0.3	owners and investing firms pay; the poorest gain	medium
Balanced policy	+10.6	+0.5	+0.8	−0.9 %	−0.1	−0.2	everyone pays a part; the effects on poverty and inequality nearly cancel	medium

Budget effect: the package's own effect on the budget balance, added up over 2028–2045, before the EU rule responds; positive brings the budget money, negative costs it. Debt, revenue and the poverty rate: difference from the base path in percentage points. Economy per head: difference in per cent. Inequality: Gini index points. Confidence: how sure the direction and size of the result are, given the assumptions it rests on.

The table says three things. First, the fall in funds can be covered, and from any base. All three tax packages bring the budget two to three times what the fall takes. The question "whether" is answered. Second, what a tax package buys under the rules is not a lower debt in 2040 but a smaller forced consolidation in the 2030s, and that is a real gain: cuts the rule would otherwise make in general services, support to the economy and education do not happen. Whoever reads the debt column alone will conclude that taxes do not help; the budget column says what they do. Third, the bases differ in who pays, and that is a decision, not a calculation. Labour taxes are progressive but feed the shadow economy and burden future pensioners. Consumption taxes are simple and durable but the poorest pay. Capital and property taxes are the fairest and give way the least, but pay with investment. The balanced policy is not a compromise that evens everything out for free. It brings in almost as much as labour taxes do, and its effect on poverty and inequality is small because the three bases pull in opposite directions.

For a decision-maker two conclusions follow. Revaluing the cadastre and the property tax is a base that does not give way and lowers neither the incomes of the poorest nor, unlike the profit tax, investment: that is where to start. And the real decision of the next five years is not on this page's packages at all but on the base path. The rule will cut about 3 points of GDP somewhere. Choosing where, and whether some of it is revenue instead, is better done in 2027 than left to the corrective arm in 2030.

What to keep in mind

The level of debt is a point under a band, and the band is about the labour market. Debt in 2040 lies between 59 and 69 % of GDP depending on two judgements with no Latvian record behind them: how easily firms replace workers with equipment, and how long they take to open the posts they want. A third, how many of the posts that departures leave are advertised again, moves it less. Together they set how deep the 2030s slump is. The ranking of the tax bases and the direction of every effect in the table hold across the band; the level does not. After 2040 the base path recovers: debt is 56 % of GDP in 2045 and the deficit 1.5 %, and those years rest on more assumptions than the years before, which is why the page prints 2040.

Rail Baltica's figures are from the public record, not from the budget law. The line's remaining cost and the European Union's share of it are what has been publicly reported, and they are what the state's share of 0.6 % of GDP a year rests on. The model builds the line from 2026. If it were cancelled from 2026, the budget would save 2.8 points of GDP over 2028–2045 before the rule, not the whole state bill, because the construction is also output and revenue; debt in 2040 would be 1.6 points lower, and by 2045 2.3 points higher than on the base path, because the rule would have cut less in the 2030s. If the European Union paid for all of it, the saving before the rule is 2.9 points and the economy keeps the line. The other committed item is the gap between what the budget's functions explain and what the record shows: about two points of GDP a year of spending that the model carries as a level from 2026 on, because the record says so, but does not attribute to a function.

The model's figures are read at the start of each year. Its 2025 deficit, 1.0 % of GDP, is the position at the start of 2025, while the recorded 2.5 % is the whole year's; the same holds for debt, 44 % at the start of 2025 against 47 % at its end. From 2026 the model carries the budget's own plan, so read the lines from 2026. End-2026 debt in the budget plan is about two points higher than the model's, because the plan starts from its own, higher, 2025 figure.

The consolidation is the rule as written, not a programme anyone has adopted. The model applies the corrective arm mechanically: savings adopted at a fixed pace once the structural deficit passes three per cent, spread across the unprotected functions, unwound slowly once the deficit is back under two per cent. A real government would choose where, when and whether some of it is revenue; it could also use the framework's flexibility for defence. The size of the consolidation is reliable; its shape and its timing are a convention.

How strongly the tax base gives way has not been measured in Latvia. How many envelope wages each point of labour tax creates, and how many undeclared transactions each point of value added tax, is taken from studies of other countries and from the Baltic and central European panel. There is also a chance that the model counts the same envelope wage twice: once as lost tax base and once as growth of the shadow economy. If so, the labour tax penalty is half what is shown, and the order of the bases does not change.

Who pays, the model sees in five income groups, no finer. The poverty and inequality rows are calculated by spreading each tax across fifths of the population by income. They give the direction and the rough size; they do not say what a particular family pays.

The model does not replay the crisis years. The recorded deficit and spending series run far from the model's line in 2009–2012 and in the pandemic years 2020–2022, because the model replays neither the crisis nor the pandemic. On the level of debt the fit is good throughout.

The 2028 start is an assumption. Every package starts in 2028 and reaches full strength in 2030, the same year the rule's savings start to bite. A later start shifts everything to the right; the relations between the bases stay the same.

How will the state pay for defence and care when EU funds shrink?

The short answer

Why it matters

What drives it

What can be done

Labour taxes

Consumption taxes

Capital and property taxes

Spending reallocation

What the packages do

Assessment

What to keep in mind