When Warren Mosler opines here that where government first spends into the economy it is employing all the people who receive its money, as it is spending into the economy because it wants something in return, it seems plausible.
He suggests that government expenditure amounts to employment and sometimes – at least, if one takes a very broad interpretation of employment, it will do. When government obliges repayment of some of this spending by law (or maybe force) in tax, then without it, he seems to suggest, we all become unemployed.
Really? In the past taxes might have been obligations and so duties such as road mending, bridge maintenance or horse transport might be required as tax. But when it was not required that didn’t make you unemployed. That was not your employment! That was an obligation. When the King’s wardrobe issued tallies acceptable as payment of tax they were indeed buying clothes for the monarch but the recipients did not produce clothes for the monarch alone. So the suggestion that a job guarantee is inherent in government taxation seems, to me, to be false.
When the Anglo-Saxons paid their Danegeld there was no employment attached. Until Pitt started income tax the revenue raising taxes were duties on imports – and indeed exports (particularly wool) so these were not general taxation in any modern sense and ended up, Poldark, or Daphne du Maurier style, simply as a smugglers charter, which became so widespread it could be argued that avoiding tax was a system of employment..
Furthermore, these days if I get a pension I have a government income without being employed and I still pay tax (although depending on the actual level of pension income, this does not necessarily mean income tax). I think, therefore we can conclude that there is nothing sacrosanct about a job in this context and the system would be just as effective with a Universal Basic Income.
It may be a plausible scheme for automatic counter-cyclical government expenditure but trying to dress up a justification for the Job Guarantee as a historical given is distinctly suspect.