Brian Goldstein writes in the comments:
1. Do you have any estimates available about the processing cost of a revenue neutral sales tax regime? i.e. Would the IRS spend only 40 cents? or 30? or 50? I know you’ve guessed that the smaller N-size of businesses means that they would realize some savings, I’m just curious to know how much.
2. Do you really think that revenue collection inefficiency is a dead-weight loss of such priority it’s worth busting up a nearly 100 year-old established system? I imagine the transaction costs alone would in the medium term be daunting, and rather than shrink the market for tax professionals, just shift the market from individuals to corporations. Corporations don’t just pay on what they sell, they pay on what they own – or not pay on what they could justify not owning. We’d still need an elaborate system of record keeping, checking, statutory provisions, and enforcement to account for those revenues.
First, as I’ve noted, from a theoretical level the costs of enforcing a sales tax regime are comparatively less than enforcing an income tax regime. Here’s the reasoning; let’s begin with an analysis of the costs of enforcing an income tax regime:
1. There are record-keeping and processing costs to individuals and corporations, since they have to keep records of income generated and calculate their own tax liabilities. Here too is included the costs of paying professionals to do the work for you. These are systemic costs borne by everyone who fills out a tax form, even if they don’t pay taxes at all.
2. The cost of enforcement. Comparatively it is far more attractive to enforce a sales tax than an income tax, if only because you have to monitor fewer entities. Enforcing income taxes means having mechanisms to audit, prosecute, and collect from the entire population of citizens who owe income taxes plus the entire population of companies that owe income taxes. Enforcing sales taxes means that we only have to have a viable enforcement mechanism for the entire set of companies.
3. Filling the gap: Income taxes are easy to cheat on. People who receive a large portion of their compensation in un-recorded cash, for instance, never pay income taxes on that income. Nor is there a viable systemic check against individuals claiming exemptions they’re not entitled to. Nor is there a viable systemic check against people hiding taxable income. And you’re only able to monitor the people who actually submit income tax forms; illegal immigrants or criminals simply don’t do this. I use the qualifier ‘viable’ because I think it’s clear that the system’s enforcement mechanisms are necessarily and woefully incomplete. Here is Steven Dubner at Freakonomics concluding that a system like ours where we lose tax income from a result of human mistakes in calculation plus willful obfuscation (cheating) is inefficient.
There are more arguments on this front but I think I have the major ones. In comparison to income taxes, sales taxes are vastly superior:
1. The systemic costs of record-keeping and processing costs involved in a sales tax regime are far less onerous. For one, they happen automatically at the point of sale, where most record-keeping is electronic. The tax, which is a percentage of the sale, is easily computable. Oh, and for Missouri specifically, or any state where there exist both sales and income taxes, the cost of switching to an exclusive sales tax regime is cheap, since the mechanisms for collecting and monitoring sales taxes are already in place. Businesses face little to no increased costs for compliance.
2. Enforcement is vastly easier. Instead of monitoring the entire set of people and corporations who owe income tax, all we have to do is monitor the entire set of corporations conducting transactions. By and large these mechanisms already exist. Cheating on your sales taxes is also hard or complicated enough that it’s hard to imagine it’ll happen on s systemic basis, since it involves falsifying point of sale records on a systemic basis to generate enough revenue to be worthwhile. It’s vastly easier to catch businesses cheating on a sales tax than it is to catch individuals cheating or making mistakes on their income tax preparations.
3. Taking sales taxes allows us to tax everyone, not just the people able and willing to do their taxes. Particularly, illegal workers or black-market income are impossible to tax using an income tax regime, but everyone buys gas and goes to the grocery store. Taxing every retail transaction undoubtedly would end the de facto exemption existing for anyone who doesn’t pay income taxes but should.
To answer Mr. Goldstein on point: No, I don’t have the calculations showing how much money ending income taxation will save, but the argument is that there’s no reason to enforce and monitor the larger population of businesses and individuals when you could just enforce and monitor sales taxes collected at businesses (assuming revenue-neutrality at least). Think of income taxes partly as a transfer of wealth from taxpayers to non-taxpayers. The costs of expanding the sales tax regime are marginal and minimal, since they already exist and to increase the sales tax businesses just have to increase the tax rate on transactions, an easy thing to do.
Ending the income tax system might involve small, one-time costs, but by and large I can’t imagine they’re large or meaningful. And yes I think that collection inefficiency is a good reason to segregate between taxation regimes. Collection efficiency is an often-neglected parameter when considering taxation systems; we want to maximize revenues and minimize collection costs. Clearly income taxes don’t do a very good job comparatively.