Vital Signs is a weekly economic wrap from UNSW economics professor and Harvard PhD Richard Holden (@profholden). Vital Signs aims to contextualise weekly economic events and cut through the noise of the data affecting global economies.

This week: An innumerate US budget puts poor people last, and Australia is yet to have a sophisticated discussion about tax reform.

The Trump administration proposed its first budget this week. The plan contains savage spending cuts totalling US$3.6 trillion over the coming decade.

And it is indefensible on almost every level.

Trump’s budget supremo Mick Mulvaney called it a “Taxpayer First Budget”. True enough. It’s also a “Poor People Last Budget”. Mulvaney said that government should show “compassion” for the less-well-off, but “also…have compassion for folks who are paying [for] it”. Hmmm.

The Trump-Mulvaney version of compassion involves slashing US$800 billion over 10 years from Medicaid – a health care program for the poorest Americans – and cutting the Supplemental Nutritional Assistance Program (SNAP, or “food stamps”) by 29%. This is a program whose sole goal is to help very poor people – often children – eat. Last year it assisted 44.2 million Americans who, without the average US$125 per person monthly benefit, might literally starve to death.

It’s a bad budget. It’s a mean-spirited budget. It’s a budget that threatens the American social contract.

But it’s also an innumerate budget.

The Trump budget involves large-scale income tax cuts. How much revenue is lost from those tax cuts depends on one’s assumptions about how much economic activity is stimulated by them.

It has rarely been the case that such income tax cuts are self-financing in the medium term – at least when the starting tax rates are at the medium-high but non-confiscatory levels they are. Nonetheless, the Trump budget assumes these tax cuts do pay for themselves.

Of course, tax revenues overall depend on how much the economy grows, and here the administration has assumed a very bullish 3.0% GDP growth figure. That is in contrast to the latest quarter growth of 0.7% on an annualised basis, and a string of low numbers (with only one quarter’s expection) going back years.

What, I hear you ask, is going to lead to this resurgence in growth? According to the administration it will come from – wait for it – tax cuts.

Weren’t they the same tax cuts that were going to generate enough additional economic activity to pay for themselves? Yes. Isn’t that double counting? You bet it is.

Or, as former US Treasury Secretary Larry Summers succinctly put it on Fox News:

“(They’re) value judgements about the poor — I don’t agree about those value judgements, but they were elected and have a right to make their value judgements. There are optimistic economic forecasts. They’re further away from the professional consensuses than any administration in more than a generation. That doesn’t seem like the best bet, but they’re entitled to their opinion. But they’re not entitled to violate the laws of arithmetic and logic.”

This suddenly makes Australia’s budget, with it’s fanciful 3.75% wage growth assumption, look reasonable. Well, not reasonable, but at least the arithmetic is correct (although I still can’t get the bank “levy” to add to A$1.6 billion annual post-tax dollars under vaguely plausible assumptions, and I think it’s actually that on a pre-tax basis, but I digress).

And although no reasonable person thinks the Trump tax cuts are self-funding, it is worth noting that we in Australia are going in the other direction, from a higher base.

We seem to be running the risk in the Australian tax debate of thinking that raising taxation rates by a percent or two, then another 0.5%, then potentially more, has little effect because all the increments are relatively small. That’s just not true. The higher the tax rate the more powerful the disincentive is. Going from and 5% tax to a 10% tax is a less big deal than from 40-45%, for example.

Labour-supply effects, that is people deciding if it’s worth them working or working more, are nonlinear. Or to put it another way, death-by-a-thousand-cuts works in both directions.

Let’s not have an unnuanced tax debate based on false dichotomies, or ignoring the starting level of taxes from which changes are made.

Trump’s budget is bad in every way. But that doesn’t excuse us here of ridiculous assumptions, unprincipled taxes on banks, and an inability to keep straight what is pre- and post-tax.

And it doesn’t mean that a 50% marginal tax rate on the most mobile, highest contributing taxpayers does not have negative consequences.

Richard Holden receives is an ARC Future Fellow.