Replace First Past the Post with Single Transferable Vote
The 2024 general election produced a parliamentary majority of 170 seats from 34% of the vote. In 2015, UKIP received 3.9 million votes and won one seat. The Greens received 1.2 million votes and also won one seat. The SNP received 1.5 million votes and won 56 seats. These outcomes are not democratic in any meaningful sense.
First Past the Post systematically distorts voter intention. It creates safe seats where millions of votes have no effect on the outcome. It forces tactical voting. It punishes new parties and rewards geographic concentration.
STV (Single Transferable Vote) in multi-member constituencies preserves the constituency link, eliminates wasted votes, and produces broadly proportional outcomes while still allowing voters to choose between individual candidates rather than party lists. It is used for all elections in Ireland and for local elections in Scotland. Both systems work well.
The evidence is clear. FPTP served a two-party system that no longer exists. Continuing to use it is not tradition, it is distortion.
Drug policy should be based on evidence of harm, not moral panic
Portugal decriminalised personal use of all drugs in 2001. In the 25 years since, drug-related deaths have fallen dramatically, HIV infections among drug users have collapsed, and overall drug use has not increased. This is not disputed. It is extensively documented in peer-reviewed research.
In the UK, drug policy is still shaped largely by the Misuse of Drugs Act 1971, a piece of legislation older than most of the people it is applied to. The classification system does not reflect relative harm. Alcohol and tobacco, which kill tens of thousands annually, are legal. Psilocybin and MDMA, which are showing significant promise in clinical trials for depression and PTSD, are Class A.
Reform the classification system to reflect evidence of actual harm. Decriminalise personal possession. Redirect enforcement spending into treatment services, which have a far higher return per pound spent. This is not about being permissive. It is about being effective. The current approach has failed by every measurable metric for five decades.
Require all government policy to pass an independent impact assessment before implementation
We require environmental impact assessments before building a car park. We do not require evidence impact assessments before implementing policies that affect millions of lives. This asymmetry is remarkable.
The What Works centres (What Works in Education, the Early Intervention Foundation, etc.) represent good practice, but they are advisory. Government departments can and do ignore them. There is no legal requirement to demonstrate that a proposed policy is likely to achieve its stated aims, or even that it will not make things worse.
Establish a statutory requirement that all major policy proposals undergo an independent evidence review before implementation. Not a veto. An assessment. Published in full. If the evidence suggests the policy will not work, the government can proceed, but it must explain publicly why it is overriding the evidence. This creates accountability without paralysis. The public deserves to know whether a policy is backed by research or by instinct.
A universal basic income pilot at local authority scale
The theoretical arguments for and against UBI are well rehearsed. What we lack is robust UK-specific evidence. Finland ran a two-year trial involving 2,000 participants. The results showed improved wellbeing, higher trust in institutions, and no reduction in employment. Similar pilots in Canada and Kenya have produced comparable findings.
But Finland is not Britain. The interaction of UBI with our tax system, housing market, and welfare infrastructure would produce different effects. We need to find out what those effects are before committing nationally.
Fund a properly designed, three-year UBI pilot in two or three local authority areas, with randomised control groups and independent academic evaluation. The cost would be significant but bounded. The knowledge gained would be invaluable. We cannot keep debating UBI in the abstract. At some point, you have to test the hypothesis. That is how evidence works.
Protect judicial independence from political interference with a codified constitution
The United Kingdom is one of only three democracies without a codified constitution (the others being Israel and New Zealand). This means that fundamental rights and institutional arrangements exist largely through convention and statute, both of which can be changed by a simple parliamentary majority.
We have seen the consequences. The prorogation of Parliament in 2019 was technically legal until the Supreme Court said it was not. The Judicial Review and Courts Act 2022 restricted the scope of judicial oversight over executive decisions. The Rwanda policy explicitly sought to override court rulings through legislation.
A codified constitution would entrench the independence of the judiciary, define the separation of powers, and require a supermajority for changes to fundamental rights. It would not prevent democratic change. It would prevent the rules of the game from being rewritten by whoever happens to be winning.
Publish every tax relief and exemption alongside the Budget, with five-year cost projections
HMRC administers approximately 340 tax reliefs and exemptions. They cost the Exchequer over £400 billion in forgone revenue each year. That is more than the entire NHS budget. Most of them are never reviewed, most are poorly targeted, and many have been in place for decades without anyone checking whether they still achieve what they were introduced to achieve.
Entrepreneurs Relief (now Business Asset Disposal Relief) was estimated by the IFS to have cost £22 billion over ten years with no meaningful evidence that it increased entrepreneurship. Patent Box costs roughly £1.2 billion a year and the evidence on whether it stimulated UK research rather than just relabelling existing activity is deeply mixed.
Require the Treasury to publish a full register of every tax expenditure alongside the main Budget document. Each entry must include: its original justification, its annual cost, its distributional impact, and an evaluation of whether it is meeting its stated aims. Any relief that has been in place for five years without evaluation should be marked as under review and subject to automatic sunset unless Parliament actively reconfirms it.
Tax expenditures are spending by another name. They should be scrutinised like spending. At the moment they are not.
What do you think?
Vote on individual policies, adapt them, or write your own alternative. Good ideas should be tested.