On tax deductions for self-education, or #ScrapTheCap
The observation that “doctors will have to pay more” is not an argument. The AMA should find a better one.
The classic lens by which people judge tax changes is in who becomes worse off. This is flawed: tax distributions always leave someone worse off. If we were afraid of that, we would never change anything. Regrettably, that doctors will incur costs is all the Australian Medical Association (AMA) has been able to come up with in its fight against a proposal to impose a $2000 cap on tax deductions for self-education expenses.
Especially here, the argument that things will cost more for some is hard to have sympathy for. Those who will lose most are in well-off, high-prestige professions like doctors and architects. While education is expensive, these people aren’t exactly poor.
It’s also not enough to say that it’ll create a disincentive. Needing to spend more obviously will, but that again is just relative to the status quo—an argument for never changing anything.
Distortion and externality
The question to ask is not who will lose from the change, but what tax structure would bring about the “right” level of spending on self-education. And you can’t just say “spending more on education is always better”: unless you think we should spend everything on education, that’s self-evidently false. There are two main elements to consider:
- What would least impact the decisions people make compared to if tax wasn’t a relevant factor?
- Are there any benefits incurred in self-education that aren’t reflected in the price paid by the individual?
In economic jargon, the first is “distortion”, and the second is “externalities”.
The reference points in italics are important. Obviously if you remove a tax deduction, people will be respectively less inclined to spend there. But it’s circular to say that the change is necessarily bad, since that assumes the status quo was good. It’s also not enough that doctors’ patients benefit from the training. They obviously do, but that’s built into price signals: more knowledgeable people get paid more.
This means you can’t just say, “people will spend less on education” as if that’s bad. You have to ask what is least distortionary (after externalities). Unfortunately this question is far too difficult to address in a sound bite. Distortion-wise, the principle behind the deduction goes like this:
- We tax everyone on their income, which by default is not distortionary.
- Self-education is money spent on being productive to others, not spent on oneself. So it is incidental to how they earn their income, so it’s not really income.
- When you apply income tax to this, you distort incentives away from spending on “factors of production” (so to speak) towards end-user spending.
- Therefore, funds spent on self-education should not be subject to income tax (i.e. should qualify for deduction).
The most tenuous premise is (2). In order for you to fully believe it, you have to believe that the professional does not gain from the education. But the education has the almost-direct purpose of increasing the professional’s income (this is a requirement). So you would have to believe that this doesn’t truly count, and also ignore that people often want this education and the stimulation (both present and future) arising from it.
This leaves externalities. Is there a benefit to society that wouldn’t be taken into account by a doctor when deciding to undertake that study? Of course, many doctors are remunerated by the state, not patients. Also, doctors don’t really decide whether to do this study; they just do. These facts complicate the question. I don’t have an answer, but it’s worth noting that we subsidise education in general in recognition of its social benefit. Whether the deduction is a good means of correcting this externality is questionable, though not implausible.
The social benefit we recognise in education also raises another quirk of this tax deduction, coming from its very definition. Not all education qualifies for a deduction: even if it relates in “only a general way” to your work, it doesn’t count. So those who advocate it because “education is valuable” should at least consider whether the deduction is unduly discriminatory.
To qualify, the course has to “skill or specific knowledge required for your then current work activities”. This disadvantages professions where one’s value lies in a broad base of knowledge not always strictly relevant to the job at hand, e.g. those heavily involving innovation. (Disclosure: I’m a telecommunications engineer.) Also, by biasing towards one’s current work, it distorts incentives away from retraining to fill a current gap in the labour market, thereby (further) reducing the reallocation of resources to skills in high demand.
Some better arguments
There are viable grounds to argue that taxing self-education is distortionary, or that there are externalities not otherwise accounted for. The AMA could also be using other arguments, for example, that their extensive training is in effect mandatory, so the state shouldn’t be taxing them for fulfilling a requirement.
That’s not what they’ve been doing. Their strategy seems to be impressing on people that doctors spend a lot on training, and that a tax deduction is a requisite for recognising its value. The first part is true but irrelevant. The second part is false. The state can have many ways of recognising the value of an activity; it may even decide it is valued enough by its citizens not to need a special push. As it stands, the “Scrap the Cap” campaign reeks of self-interest alone, not unlike the various oppositions to the quagmire of deductions in America.
This is regrettable. The AMA should be expected to campaign on behalf of its members—that is an association’s job. But by framing the issue primarily in cost to practitioners, they do their members and the public a disservice. Doctors rightly have a reputation for caring about the impact their work has on the community. I hope the campaign will begin to reflect this soon.