Quote from OldTrader:
it will cost you an extra $.50 per share.
Surprisingly, it easily could, even without any drops in volume. In the interest of fairness, because it is always likely to be the most powerful of arguments, I've tried to estimate the range of possible outcomes (market impact) of introducing a 0.25% transactions tax. One of the most optimistic scenarios resulted in a bid/ask spread of 0.05, but there are also more pessimistic ones, when not only 0.50% extra cost has to be borne by the economy (regardless of the exemptions), but market volume drops by half. I used a representative share price of $100, and a marginal costs estimate (brokerage commisions) from an industry-leading discount broker. Other costs of doing the market-making business were assumed to be zero, i.e. that retail traders have only variable costs, i.e. brokerage commissions.
In every competitive market, traders are simply price-takers and the price reflects their marginal cost of doing business. So a mutual fund manager buying a popular stock (such as the S&P 500 ETF priced at $100) has to pay 1 cent per 1-lot purchase of $10k worth of this stock, to be serviced by traders in today's competitive markets. When does this cost come from? It arises because traders have to pass their costs onto their client - the fund. We are not talking brokerage commissions now, which are likely much lower for institutions, we are talking the immediate difference between the purchase price and the post-purchase market value (to be reported to the fund's shareholders), i.e. the so called bid/ask spread. If the price does not change at the end of the reporting period, the fund will report a 1-dollar loss (1 cent lost on a 100-share lot of $100-shares), because the mark-to-market rule applies and the SELL price (market bid) is lower by 1 cent than the fund's purchase price (market ask). Is this cost of doing business high? 1 part per thousand, I'd say pretty low, so the fund manager can charge her clients less in management fees than he would in other, less competitive countries.
Let's just assume that there are no other costs of doing the trading business, and the currently observed 1 cent difference between the BUY and SELL prices of most liquid US stocks is so low only thanks to the round-trip brokerage commissions for the retail trader being as low as 1 cent (1 dollar per trade, for lots of 200 shares or more, see e.g. the Interactive Brokers commission schedule). 1 dollar paid on each 200-share trade (2 dollars both ways) is exactly equiavalent to a 1-penny spread between the stock market prices (2/(200*$100) = 0.01%= 0.01/$100).
If you raised brokerage commissions, e.g. 10 times, it would no longer make sense to bid the stock to buyers at 99.99 and offer the stock for sale at 100.00, because the currently available profit of 2 dollars would be insufficient to meet the increased round-trip cost of 20 dollars (10 per BUY and 10 per SELL transaction). Just like an increase in the Value Added Tax would be always passed onto the consumers (added to the purchase price), traders would have to reluctantly add the raised cost to their calculation of the offered price, simply to break-even.
Reluctantly, because they know how competitive the market is, and that they are faced with a very elastic, almost flat demand curve (has any regulator ever tried to check its elasticity before considering this tax?). In such cases, raising quoted prices just a tiny bit immediately stifles demand disproportionately more (to almost zero). No-one would ever come to the trader's corner shop to buy stocks if she raised her offered stock price to 100.05, where the competitor on the other side of the street offered the same standardized stock at only 100.01. Moreover, no-one wants shares spread 0.05 cents apart even if there aren't any more competitive offers... no transactions are made in periods and on issues when and where only the market making firms have the monopoly to serve the client.
In a competitive market, even a seemingly small (by main street standards) transaction tax of just a quarter of a percentage point could be easily larger than the traders' revenues, so it will be immediately passed onto the consumer - the mutual funds (and then to their clients, in management fees and/or lower returns). After the 0.50% tax is imposed (the total tax impact is always *twice* the advertised 0.25%, because the fund has to sell eventually to lock-in gains), the fund manager will have to pay 0.5% more for each round-trip stocks transaction (which she makes on average at least once on the entire stocks portfolio in the process called 'rebalancing' - see the annual turnover column in Morningstar reports).
You quickly say: retirement accounts will be exempt, so that's not a problem. But that would be equivalent to exempting from VAT those grocery stores in which only pension fund manager spouses (here: husbands) are allowed to shop (and exemptions for market making firms are not enough - see below). Can you selectively exempt from the transaction tax some stock market traders, allowing them to keep on quoting to some preferred 'good cause' clients the current tightly-spread prices such as 99.99/100.00)? No you cannot do that, because traders are anonymous and market access is more or less equal, and there are no preferred buyers when it comes to price (even big players using dark pools have to transact at the same price as all other market participants). So your retirement account will have to pay that 0.5% tax - not directly (exemption will apply on that level), but because the prices will no longer be just 1 penny apart. They will be 0.5% apart... because you have raised the cost of doing business to the liquidity providing traders. They can no longer afford to quote you a penny-wide spread, because the quoted offer price has to include your new 'Tobin friction' - the round-trip cost of 0.25% paid *twice* (for buying and selling) on those 100 shares worth $100. In total $50=2*25=2*0.25%*100*100 dollars. On a 200-share lot, the cost of the new tax would be a round sum of 100 dollars. When compared with the today's round-trip brokerage commission cost (which we assumed liberally to be the only cost of doing the trading business) of just 2 dollars (as computed before), you see that the bid/ask spreads would have to raise 50-fold for the traders just to break even. Hence the 'outrageous' 0.5 spread amount computed by the OldTrader before me. So in future, the mutual fund or the retirement account investor, punished equally with everyone else by this tax, will see the following prices:
99.75 - 100.25 (i.e. the round-trip tax rate apart),
instead of today's:
99.99 - 100.00 (i.e. the round-trip brokerage commission rate apart),
and buying this stock will immediately offset the fund by -0.50%. This will be the new mark-to-market return reported to the mutual fund shareholders if the price did not change at the end of the period, 50 times worse than the current -0.01%.
And please do not think that exempting maker makers would improve the situation - these firms have large overheads (light-speed technology has to cost after all), so when they are alone, unhindered by the competition, e.g. at 5 AM at night, they offer our stock (which is most liquid in the world, incidentally) at 100.05, five times more expensively than the retail trading community would.
So here we have a range of estimates of the adverse impact of this tax on the retirement plans and funds:
- 5 cents per $10k (-0.05% loss from today, i.e. 20% of the maximum tax revenue of +0.25%, earned assuming heroically no change in aggregate volume, and that the monopoly power of the market makers will not be exploited more than it is today, and they will keep quoting prices as they do today in periods without competition),
- 50 cents per $10k (-0.5% loss from today, i.e. 200% of the maximum tax revenue, which will happen if market makers join the marginal cost payers, i.e the taxed traders, with the other heroic volume assumptions maintained as above),
- 50+ cents per $10k (-0.5% loss from today, but with an additional -50% drop in volume, i.e. a 400% ratio of the economy-wide cost to the projected tax revenue - now raising only half of the projected amount; computed without any heroic assumptions made, i.e. letting the market makers behave as rational agents, joining the marginal cost payers, and using the Swedish case to estimate the extent of the decline in the stock market volume),