President Trump’s proposed budget justifies its potential $1.5 trillion increase in the Federal deficit over ten years by claiming it will increase the anemic U.S. growth rate (thereby paying for itself). There can after the third quarter figures released Friday be no question that the U.S. economic growth rate has increased under Trump, with his deregulation (or simply ceasing to pile on additional regulation) the main spur to boosting an economy that had been burdened with non-market interest rates and silly leftist regulatory theories. But Trump’s proposed tax changes are a different matter; they are by no means clearly supply-side in their likely effect, and thus may simply deprive the Federal government of much-needed revenue.

To get a significantly increased growth rate from a tax change, it is necessary that it have a positive supply-side effect. A flat tax cut of $500 per citizen has no supply side effect because everybody’s return from a marginal hour of extra work is unaltered by it. It affects only the demand side, by increasing consumption (and has a corresponding dampening effect from the addition to the budget deficit it causes). By and large, Keynesian multipliers by which a tax cut or spending boost produce knock-on effects on overall output are very small and may even be negative when the additional government borrowing is taken into account.

The supply side paradigm can also be applied to corporate tax cuts, but here the incentive effects are generally more complex, since an increase in corporate net income may simply raise share prices, without generating any extra enthusiasm for domestic capital spending, let alone increasing the wages of the workforce.

To begin with the proposed corporate tax changes, Trump proposes to cut the tax rate to 20% from the current 35%, but make up some of the revenue loss with a low rate of tax on international earnings and accumulated assets. This will make a gigantic hole in the budget, though the one-off tax revenue may postpone fiscal disaster for a year or two. Trump’s proposed corporate tax structure is certainly better than the alternative proposed by Chamber of Commerce types – making tax truly territorial, thereby making all non-U.S. earnings tax-free if they can be routed through tax havens. That’s an obvious strategy for economic disaster for the United States, if not for Third World countries to which investment is re-directed.

The incentives Trump’s corporate tax change will produce are rather complex. Certainly, it will eliminate the habit of keeping all overseas earnings parked in a tax haven; there will be no additional disincentive to bringing them home since tax at a low rate will already have been paid. However, the incentive to re-direct earnings out of the U.S. will remain, at least at a lower level; if foreign earnings are taxed at 5-10% and domestic earnings are taxed at 20%, it will still be more profitable to locate earnings overseas.

A truly global tax system, in which the U.S. taxes earnings worldwide, with appropriate allowances for credit risks and losses, would be much more effective and yield more revenue for the U.S. Provided the global tax rate was relatively low, at around 20%, it would not incentivize companies to move abroad. Of course, the extra tax costs and the reduced opportunities for game-playing are why the multinational companies are universally opposed to such a system. But for medium-sized and smaller businesses, which do not have significant operations overseas, such a system would be much more favorable – and it would also yield much more tax, at great benefit to individual taxpayers and the economy. It would also remove the perverse incentives for tech and other companies to outsource production to East Asia and their intellectual property to a tax haven.

Proponents of Trump’s tax plan claim the lower tax rate on corporations would have a massive supply-side effect on domestic business, by increasing its after-tax profitability. However, profitability does not currently seem to be the problem for U.S. publicly owned corporations. Corporate profits have been hovering for several years at record levels in terms of GDP, comparable to their level in 1929. If high corporate profits spurred innovation and new investment, we should be seeing a massive investment boom. Instead, we are seeing a boom in stock buybacks, the world’s least productive way to spend corporate money. Stock buybacks boost stock prices in the short-term and thereby the value of management’ s stock options, but leave the company over-leveraged, under-invested in real fixed assets and highly vulnerable to even a modest downturn in the business.

The very expensive reduction in corporate tax is thus unlikely to do anything much beyond pushing up an already overvalued stock market. There is certainly no reason why it should give a $4,000 raise to every worker in America, as Trump has claimed. If the reduction in tax rates does not increase capital investment, as seems likely, it will neither increase employment nor raise wage rates, except for financial engineers. The reduction in outsourcing, and a possible increase in foreign investment in the United States, attracted by the lower tax rates, are both likely to be helpful, however. Thus, we should not dismiss the corporate tax cut altogether, but given its $1.5 trillion estimated cost, it does not look to be good value for money.

On the individual side, one clearly anti-supply-side tax provision is that eliminating the deductibility of state and local taxes. The arithmetic here is simple; for a Californian paying the top 13% state tax rate, the top marginal rate will not decline with the reduction in high rates generally but will increase by about 3% net to around 51%, including the Medicare tax rates payable to fund a modest part of Obamacare.

There is thus a negative supply side effect from the elimination of the state and local tax deduction; anybody affected by it will pay income tax at a higher marginal rate. Naturally, there will also be an effect forcing the well-off to flee New York and California and migrate to states without an income tax; the economic effect of this is probably neutral. In addition, raising the top rate of income tax to create a “millionaire’s tax” as both Trump and his Svengali Steve Bannon have suggested, would make the problem much worse. An increase to 44% in the top Federal tax rate, combined with abolishing the state and local tax deduction, would make California and New York top taxpayers pay a marginal rate of around 60%. At that level, the top incentive to them becomes tax avoidance, and wealth is actively destroyed.

The negative supply side effect of abolishing the state and local tax deduction would not be paralleled by a similar effect from eliminating the other possible big deductions, for mortgage interest and charitable donations. The mortgage interest deduction goes mostly to the well-off (unlike that for state and local taxes, which goes to many people with modest incomes – including myself). It incentivizes excessive leverage in-house purchases and raises house prices, especially in wealthy enclaves such as New York and San Francisco. As such, its economic effect is mostly negative. Abolishing it would make the National Association of Realtors squawk, but be economically beneficial on balance, re-directing investment to more productive sectors.

The charitable tax deduction is an even better candidate for abolition, as this column has noted before. It is the principal means by which the ultra-rich avoid paying the same rates of taxes as the rest of us, and as the Clinton Foundation amply demonstrates, it creates not productive economic activity, nor charitable munificence, but the most appalling scams. The fact that there has been no serious attempt to abolish the charitable tax deduction is a sign of the adoration for it among the donor class.

The economy, however, would be enormously benefited by its abolition (apart from the extra tax revenue), as most of the 5-6% of GDP generated by mostly spurious charities would be redirected to productive ends.

At the very least, set a cap of say $100,000 on deductions as a whole. That would allow people to choose whether to deduct state and local taxes, mortgage interest or charitable deductions, but only a limited amount of all three combined. This, above all other reforms, would make the very rich pay the taxes they should, while having if anything a positive supply-side effect.

Trump’s proposals have only a modest supply-side benefit on the corporate side and a substantial supply-side cost on the individual side. Only deregulation and interest rate increases (making capital allocation more optimal) may restore U.S. economic growth; his tax changes certainly won’t. This column sets out a much better approach, under which the rich will pay their fair share of taxes and supply-side effects will be truly optimized.