At the same time as the tunnel decision was announced, a new report suggested that ministers will be left with an £8.5bn black hole if they go ahead with the route. I haven't had time to read the report itself and it was commissioned by 17 councils along the route, so comes with a health warning. According to the Telegraph the report reveals 'that the benefits of the scheme to the taxpayer could be as little as half the costs'. This seems an odd point to emphasise, because the fact that the costs exceed revenues has been known for a long time (for example, in March 2010, I noted 'the second certainty is that any new route will not be commercially viable and will need large government subsidies'). It's also the case that lots of government expenditure fails this test because it's standard to take in to account the benefits to tax payers as well as the revenues that go to the exchequer. The fact that total benefits are likely to exceed costs is not in (that much) doubt for HS2. The point is that these total benefits don't add up to much compared to other things the government could spend money on.
The other point central point raised by the report directly addresses the issue of total benefits and concerns the government's projections about the level of demand. Figuring out the level of demand involves making predictions about growth and predictions about how quickly rail demand will increase as the economy grows. The first of these numbers is certainly open to debate (DfT claims that the figures it uses - the 'elasticity' - are conservative; this is certainly not an opinion shared by all experts in this area). You would imagine that the second number - the growth rate of the economy over the period - might also usefully be revisited (given the Autumn statement last week) but I haven't yet seen anyone do this.
The growth rate in passenger numbers matters because HS2 makes journeys faster and relieves capacity constraints on the existing network. The benefits from higher speed depend on the number of passengers that will use the service, if you project smaller numbers you get smaller benefits. The capacity constraint depends on how fast passenger numbers grow relative to capacity. Figures released this week have thrown some doubts on that problem as well, suggesting that peak time trains are only a little over half full. Again, these figures come from opponents so are subject to the usual health warning but they do raise the issue of whether other options (e.g. pricing structures, existing capacity) could be used to deal with this problem. In addition, in the Eddington study there were capacity constraints all over the transport network (road and rail) by 2025 so the case for HS2 is not unique in that sense. Finally, note that the easing of capacity constraints is also well captured in the traditional cost benefit analysis (providing that passenger numbers aren't over estimated) and as I have said before that case looks weak relative to alternative transport investments.
