Talk:Automated Payment Transaction tax

Progressive/regressive nature of APT
My edits pointing out the regressive nature of the APT seem to be controversial, and have twice now been removed.

I think it is helpful to point out the following:

If an earner makes either $20,000 or $2,000,000 through wages (or dividends, or capital gains), the APT tax would be the same, since the rate does not scale in proportion to the amount.

This seems to show that the APT tax is a flat tax, and thus regressive. It is not obvious to me how this example can be wrong. Shlaer (talk) 04:33, 20 November 2012 (UTC)


 * I am new to editing so I think I put my talk comments on the wrong page. I hope this is the correct place for them: I hope we can use this forum to come to some agreement on what is an appropriate description of the APT tax plan. The first area of disagreement appears to concern the issue of whether the APT tax is progressive or regressive. The article cited in my version of the text contains empirical evidence based on the federal reserve's Survey of Consumer Finances that demonstrates that the most wealthy deciles undertake a disproportionate volume (in dollar terms) of transactions. Progressiveness is introduced not via the rate structure (which is indeed flat) but uniquely through the tax base since the wealthy undertake a disproportionate share of financial transactions (eg stocks, bonds and derivatives) You appear to disregard this evidence and make the following argument: "However, it is unlikely that the number of transactions per dollar spent is indeed proportional to income." What is relevant here is that the paper demonstrates that the dollar volume of transactions per dollar of income or per dollar of wealth does indeed increase with the level of income and wealth. See Figures 2 and 3 in the cited text. You go on to give the example: "For example, regardless if an earner makes $20,000 or $2,000,000 per year through wages or dividends, the tax rate paid by the earner would be the same, since the number of financial transactions occurring does not depend on the amount earned." Of course, the tax rate is a fixed flat rate, but the tax burden on the $2,000,000 earner will be much higher because that earner undoubtedly holds and trades a volume of stocks, bonds and other assets and whenever these are exchanged, purchased or sold, they must pay the APT tax whereas the $20,000 earner probably has few if any assets to trade. I look forward to your response and if we can agree, perhaps we can address some of the other issues on which we seem to strongly disagree. Kgelfman (talk) 21:28, 20 November 2012 (UTC)


 * Yes, I agree the most important disagreement is the progressive vs. regressive issue. I looked at Dr. Feige's simulation results, which you cited, and it appears his methodology does not include how people would behave after APT were implemented. It would be reasonable to assume that if transactions were taxed, then they would be avoided whenever possible.   Furthermore, if the hypothesis is merely that the wealthy will engage in a disproportionate volume (in dollar terms) of financial transactions, this does not mean that the wealthy engage in a disproportionate volume of transactions per dollar spent, because the wealthy also spend more.  In order for APT to be progressive, the wealthy would have to engage in more financial transactions per dollar spent, not simply more financial transactions.  This would be the case if a more wealthy individual rebalanced (or turned over) their portfolio more often than a less wealthy individual.  But it would not be the case if the more wealthy individual simply owned more stocks.  As an example, imagine that over the course of one year, the more-wealthy individual bought $1,000,000 in stocks and sold stocks for $1,100,000, thus earning $100,000 in capital gains.  The tax on this would be the same rate as for a less wealthy investor who bought $10,000 in stocks and sold all of it for $11,000, earning only $1,000 in the same period.  This is quite similar to my example with wages or dividends.  (An unfortunate and highly regressive scenario would arise if some investor lost money through a capital loss, say by buying $50,000 in stocks and only selling them for $49,000.  The individual would not only lose $1,000 in the trades, but also have a tax burden on the $99,000 in transactions.)


 * Then if we assume that the vast majority of individuals earn money through some combination of wages, dividends, or capital gains, the tax rate would be flat. Including capital losses makes the tax even more regressive than a flat tax on income. Shlaer (talk) 23:25, 20 November 2012 (UTC)


 * It seems the disagreement arises due to a misunderstanding of how the APT tax works. It is a flat rate tax on the total volume of transactions undertaken. Let me take your example to show the progressiveness of the APT tax. Your person A making $20,000 per year spends his entire income ($20,000) and owns no assets. For simplicity let the APT flat rate be 1%. He will pay 1% when his income is paid ($200) and another 1% when he spends his $20,000 (another $200) for a total tax burden of $400. His income tax rate = $400/20,000 = 2%


 * Individual B makes $2,000,000 and say consumes $1,500,000 and adds $500,000 to his portfolio of stocks which say amount to $10,000,000. His 1% APT tax paid on his income, ($20,000); on his   consumption ($15,000) and this years stock purchase ($5000) amounts to $40,000. However we know that people who ear $2,000,000 per year also have a portfolio of assets. Say B’s has a portfolio of stocks worth say $10,000,000, which after the APT tax is imposed, turns over once a year. This gives rise to stock sales of $10,000,000 and stock purchases of $10,000,000 equaling $20,000,000 in transactions which when taxed at 1% gives rise to a tax of $200,000. B’s total tax payments are thus $40,000 + $200,000 = $240,000. His income tax rate is $240,000/2,000,000 = 12%...  If A pays a rate of 2% and B pays a rate of 12% then clearly the tax is progressive. Note that if B only makes transactions of $10,000 from his asset portfolio, his total taxes will still amount to $40,100. B’s tax rate is then 2.005%. Since the tax rate rises with income, the APT is progressive. Kgelfman (talk) 22:38, 21 November 2012 (UTC)


 * You forgot to include investment income. Individual B has a stock portfolio worth $10,000,000.00, and it will generate (say) 10% annual return, i.e., $1,000,000. The tax on this would be $200,000 and so an effective tax rate of 20% on investment income, regardless of the amount. This sounds like current law, except that investment losses are also taxed, so investment is now a highly risky proposition, and leveraging is impossible. No investor would turn over their entire portfolio, as this would be just like giving away $200,000 for no good reason.  Why pay 20% tax when you can pay 2% by leaving your assets invested in one mutual fund forever?


 * But regardless of this, there is no reason to assume that individuals making $2,000,000 per year will additionally have a huge investment portfolio which is generating almost their entire tax burden. I think we should first resolve this issue using the model I originally suggested, and not assume the wealthy have new transactions not justified by the model I presented. Shlaer (talk) 05:55, 24 November 2012 (UTC)


 * I did not forgot investment income. Indeed individual B may earn $1,000,000 on his ten million dollar portfolio but that of course is already part of his two million dollar income. We already figured the tax on that income so I have no idea where you get a tax of $200,000? Making up examples like this is fruitless. Feige's paper empirically examines the relationship between transactions and income and transactions and wealth and finds that the richest people also undertake a disproportionate volume of transactions. That finding is sufficient to show that the APT tax will be progressive.


 * Investors can undertake any investment that they hope to be profitable after paying a tiny "brokerage fee" to the government. Just as your broker does not share in your gains or losses, neither does the government under the APT tax. The government simply charges you a brokerage fee whenever you exchange goods, services or assets.
 * Finally, you never presented a model. You presented a flawed example. I think this would best be resolved by foregoing opinions based on faulty examples and instead citing professional literature that has dealt seriously with the subject.Kgelfman (talk) 20:05, 24 November 2012 (UTC)


 * Clearly we can't trust a model based on data gathered about people not subject to the APT tax, as Feige's model unfortunately does.  Can you find published research showing expected annual portfolio turnover under an APT tax?  If not, should we be using portfolio turnover to prove how progressive or regressive the APT tax is? Shlaer (talk) 00:27, 25 November 2012 (UTC)
 * The simulations from the consumer survey of finances directly refer to those consumers who would be subject to the APT tax if it were imposed. In short, your critique is misplaced. Another editor attempted to resolve this dispute by indicating that there was an issue concerning the question of progressiveness. To date, the only individual to suggest that an APT tax (or any financial transaction tax) might be regressive is Shaler who continually undoes the revisions. I have restored the neutral position taken by the last contributor in light of this ongoing controversy. Kgelfman (talk) 23:15, 27 November 2012 (UTC)
 * I think that consensus needs to be reached here on the talk page, and we should not resort to undoing each others edits. I have an important criticism of the content of the page, based on simple arithmetic. It provides needed balance for your version of the article.  If you find my criticism to be based on some error, then please explain with a simple, straightforward answer, and build consensus.  As I mentioned several times before, Feige's graphs showing positive correlation between income vs. transactions under current tax law is not germane to my criticism. Shlaer (talk) 02:27, 28 November 2012 (UTC)

WP:3O Request
Hello all, I am a volunteer who has not been previously involved in this article, responding to a request for a third opinion made at WP:3O. As you might have noticed, I have refactored the discussion above to match the standard threaded format used on talk pages here. First off, I see a history of edit warring in the article, with several rounds of reversions. This is a serious issue, and should not be continued. Consensus should be reached before making a controversial edit, and if an edit is reverted, that should be taken as evidence that more discussion is needed.

As for the actual discussion, please bear in mind that Wikipedia is not a publisher of original thought. The examples listed in the talk page above may be helpful in reaching a consensus, but the actual article content needs to be verifiable in neutral, reliable sources. Calculations performed by editors on the talk page should explicitly not form the basis of edits to the article. Particularly for politically charged issues such as tax policy, sources by partisan groups should probably not be used. There is a noticeboard where, if need be, sources can be posted for broader evaluation for reliability. VQuakr (talk) 04:58, 28 November 2012 (UTC)


 * I agree with VQuakr regarding the verifiable aspects here - the discussion seems to center around original research (the opinions of the editors involved) and not reliable sources to which they can reference. Understandably, APT probably hasn't generated significant coverage in reliable sources, so they're working with limited material.  In that case, it would be best to leave such speculation out of the article until some research presents a conclusion to support an argument.  Based on what I read, it sounds like a proportional tax as applied to the tax base.  Note I said tax base and not income, as per the definition of these terms which tends to get twisted in such debates by applying a tax on consumption, transactions, etc (which involve multi-year timeframes) to something other than the base of taxation - income.  Based on my quick read, this appears to be part of the debate, which would be mathematically inaccurate if that is the case and the base is not income.  I haven't read enough to form a position, but as stated above, it may be moot unless sources are provided that substantiate the argument.  Morphh   (talk) 15:02, 28 November 2012 (UTC)
 * Just noticed that the article says this replaces all Unites States taxes. Which caught me when I read "The APT approach would extend the tax base from income, consumption and wealth to all transactions", since the Federal tax system has no tax on consumption or wealth.  I don't believe replacing all taxes is even constitutionally possible unless each State and local municipality voluntarily implemented their own to piggy-back.  One site includes state level taxes and the other site doesn't.  This needs to be broken down a bit into how it would be implemented.  Morphh   (talk) 15:29, 28 November 2012 (UTC)

Assessment comment
Substituted at 18:09, 17 July 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 one external links on Automated Payment Transaction tax. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20120321195121/http://129.3.20.41/eps/pe/papers/0506/0506011.pdf to http://129.3.20.41/eps/pe/papers/0506/0506011.pdf
 * Added archive https://web.archive.org/web/20120321195121/http://129.3.20.41/eps/pe/papers/0506/0506011.pdf to http://129.3.20.41/eps/pe/papers/0506/0506011.pdf
 * Added archive https://web.archive.org/web/20140110060700/http://thetransactiontax.org/ to http://www.thetransactiontax.org

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 05:58, 22 October 2016 (UTC)