The Brexit debates in the UK are hampered by the problem that political experts generally have no knowledge of, or interest in, EU laws and politics; while those that do tend to be rather uninformed about British constitutional processes and political forces.

A rare exception to this is the Institute of Government, which has produced a very good analysis of the Brexit options ahead. Not one of these options would result in a second referendum. Most result in a no-deal Brexit. The five scenarios are best summarised with the following flowchart. 

The report makes the important point that it does not matter whether parliament has a majority for a no-deal Brexit. That is largely irrelevant given Article 50. When a deal is on the table, the option will be to take it or leave it. There will not be enough time to force the government to renegotiate. And, as we pointed out before, the EU would not be a willing accomplice in this game.

The report also dismisses the possibility of a new election (in a much stronger way than we would) given the powers of the Fixed Term Parliaments Act. The argument is that Tory MPs would not support it. We noted another report on an internal poll by Conservative party HQ, forecasting a loss of 25 seats if elections were to be held today. This reduces Theresa May's blackmail potential. MPs could vote against her withdrawal treaty, and in a separate vote reject a request by May for new elections. 

The report is at its strongest in its dismissal of a second referendum. One issue is time. It can't be done before March. A polarised parliament will fight over the rules of engagement - eligibility to vote, inclusion of younger voters, inclusion of EU citizens in the UK and UK citizens abroad. The problem with the three-way referendum is that it would only solve the impasse if there was an actual majority (i.e. more than 50%) for Remain. That is not the case now. And, since there won't be a deal on the future relationship in any case, MPs might be tempted to insert other options on the ballot paper - like EEA membership - so that the three options rapidly become four or five. The article also makes an important point about the EU's strategic goals: it will not allow an extension of the Brexit process unless there is clear evidence of a change of position (not through opinion polls, we would add, but through an election or a parliamentary vote). The EU would reject an extension if there was a real risk that the process would still end up in a no-deal Brexit.

"So, in practice, unless there is a clearly emerging majority (which can survive a bitter referendum campaign) for Remain, a referendum would solve nothing."