Immediately after the Scottish Independence referendum Nigel Farage complained that the UK government “made a promise to maintain the Barnett Formula whereby the UK taxpayer spends £1,600 more on every Scot than on every English person”. This kind of factoid is the sort of thing that immediately pisses people off. The problem is that it’s a nonsense figure that refers to nothing meaningful and hides far more important information.
If government spending consisted entirely of paying money directly into people’s pockets then he might have a point. Some is spent that way, of course, but, as everyone is well aware, a huge chunk of government spending pays for goods and services. There would be no point in the government paying people just to pay taxes. The government wants certain things done. For example, a large part goes to the NHS. The NHS then translates that money into a health services. The taxpayer then receives those services, not the money. The NHS provides those services by spending it’s budget on provisioning the goods and services necessary to provide health services. Ultimately the money it is budgeted ends up in the revenue stream of companies that provide it goods and services, some of which will end up in people’s pay packets and some of which will end up in share holders’ and owners’ pockets. Those share holders and owners could be anyone, anywhere. They could, and statistically are likely to be, in London. They could be in the US. The fact that the initial spending is geographically, in a sense, in Scotland has almost no baring on how the money filters down to people.
Another fact is that cost structures vary. So, say for the sake of argument, that for some reason transporting things is more expensive in Scotland. Maybe because of all of the hills. That means that provisioning of services to the public via the public sector will incur higher costs, say, to pay for extra fuel. That extra payment would show up, again, in the revenue stream of a company, in the case a multinational energy company. Or more precisely some weird holding account system somewhere so that shareholder profits can be maximised. Again, where the money ends up is a matter of topology of the economy, which has almost nothing to do with geography.
It might be, arguably, more reasonable to look at spending per head at the whole-nation level, but that figure is largely meaningless as it could be anything. Working out where money really ends up would, as far as I can tell, be perfectly doable. What makes it hard is that we would need access to data on where companies spend their money and, specifically, who all the shareholders are. That data might be available, partially, through public audit records, but we have no intention of really knowing because that might be too revealing. It’s far better to use nonsense metrics like spending-per-head.