Building on my last post on why the British Columbia provincial HST is in trouble, I wonder, even when some environmental regulation is accepted as being necessary, why environmental taxes are so disfavored. Among the federal political parties in Canada, we have the anomalous situation of the governing Conservative Party favoring some sort of a command-and-control approach to climate change regulation, while the last-century-liberalism NDP favors a cap-and-trade approach. Neither party is being honest, of course; the Prime Minister will hand out opaque sweetheart deals on a sector-by-sector basis, producing no emissions reductions, and stricken NDP leader Jack Layton promises that a cap-and-trade program will not raise energy prices which, of course, would therefore produce no emissions reductions. And the party that has championed a carbon tax, the Liberals, swooned to their lowest Parliamentary representation in recent history in the recent election.
Why is "tax" a dirty word? In my forthcoming book on carbon taxes, I argue that people harbor certain psychological biases in evaluating policy instruments. First, people have demonstrated a propensity to "do no harm" not only in their private lives, but in choosing policy instruments. The "do no harm" phenomenon, most thoroughly studied by Jonathan Baron, suggests that people prefer to allow more harm to occur by omission than they would permit to allow by an affirmative act or policy. So, for example, a Baron and Ilana Ritov study showed that there was a policy bias in favor of letting children die of the flu than for allowing children to die from a vaccination policy. How does this make taxes unpopular? I argue in my book that an environmental tax more obviously harms people than any other policy instrument. More than any other policy instrument, a voter can imagine and visualize somebody that would be harmed by an environmental tax -- it is often themselves that they can see being harmed by a tax! By contrast, a command-and-control-style regulation very effectively conceals from voters the true cost of regulation. People tend to harbor the illusion that imposition of a regulatory costs will stay where it is imposed, an illusion that Jack Layton either harbors himself or actively perpetuates for political gain.
Second, people also harbor an "identifiability bias" in favor of harming people that are less tangible, less noticeable, and less identifiable. George Loewenstein has most extensively studied this, finding that even a tiny increment of information -- the addition of an id number, without a name, face or any other identifying information -- tended to elicit more sympathy from research subjects than those without an id number. How does this cast a shadow over environmental taxes? Environmental taxes harm identifiable people, as opposed to other instruments, that harm unidentifiable people. Command-and-control, and even cap-and-trade, impose amorphous costs on people and firms that do not exist as clearly in most peoples' minds.
Finally, the famous endowment effect, pioneered by Daniel Kahneman, Jack Knetsch, and Richard Thaler, seemed to indicate that people had a much higher valuation for goods that they already had, as opposed to goods that they didn't have. Some economists have expressed this people having a higher willingness to accept than they have a willingness to pay, but from my perspective, I think it demonstrates that people have a status quo bias -- phraseology not original to me, but most useful for me purposes. And the connection to taxes? At bottom, the endowment effect is a reluctance to trade, and a tax, more than any other policy instrument -- signals a trade. The trade could well be a beneficial one, as many environmental taxes can be demonstrated to produce much more benefit than cost, but it is a trade nonetheless, and people don't like trading away from their current situation. Again, what is so glaringly obvious about an environmental tax is that improving the environmental will cost. The fact that the cost is less than the benefit is a calculation that is a second-order calculation that is dominated by the fact of the cost.