Benjamin Franklin once wrote, “In this world nothing can be said to be certain, except death and taxes.” True enough, but he missed that the magnitude of the latter is anything but certain. Savvy investors already try to reduce their taxes by investing in tax-favored accounts such as 529s, IRAs, or 401(k)s and avoiding investments, such as active mutual funds, that pay out distributions in a taxable brokerage. A number of new proposals for a blanket financial transaction tax would completely scramble this math by taxing all financial transactions.
The What and Why of a Financial Transaction Tax A financial transaction tax is not a novel idea. Financial transaction taxes have been proposed in the United States going back decades and exist in many other markets including the United Kingdom and France. In fact, the U.S. already has a very small version of a financial transaction tax. A so-called "Section 31" fee is assessed on securities sold on all national exchanges and is used to fund the SEC's operations. This fee is an extremely low rate of 0.00207%, whereas financial transaction tax proposals would increase rates nearly 50- to more than 240-fold, all the way up to 0.1% to 0.5% per transaction.
The concept of the tax is relatively simple: Tax a small portion of sales and/or purchases of financial securities in the U.S. The details can get fuzzy: what transactions count as “in the U.S.”? Are all securities taxed the same? What rate or fee is charged?
It is fair if you are wondering at this point: If we already have a small transaction fee and it successfully funds the SEC, why were multiple financial transaction tax bills introduced in the Senate last year, and why is it a part of budget and reform proposals for three presidential candidates across the ideological spectrum, including Bernie Sanders, Elizabeth Warren and Michael Bloomberg so far? Proponents of the tax say that it will reduce high-frequency trading, making the market less risky and protecting against a financial crisis, and that those traders would contribute the lion's share of the funding, minimally impacting most investors. It is undeniable that high-frequency traders would have a greater burden than most investors; nevertheless, there is a question about how much this would still affect average investors.
A financial transaction tax would impact investors in a couple of ways as the market reacts to the tax. As this would be a tax on all transactions, it would increase taxes on accounts traditionally shielded from them, such as 401(k) plans. It would likely decrease asset values because trading them becomes--even if only slightly--more expensive. There are opponents of the idea who say it could put market liquidity at risk from the reduction in trade volume. These impacts can only be indirectly estimated and involve dynamic modeling, whether it be from testing a range of responses or assessing how markets in other countries reacted.
The Financial Transaction Tax Would Affect Different Fund Strategies Differently One result that we can more directly assess is how mutual fund investment returns would decrease because of the taxation of trades within the portfolio. Although there are nuances to how the taxes are implemented, the cost would almost certainly be passed on to shareholders. So the key questions are: To what extent will this tax affect different kinds of investment strategies differently? What does that mean for a financial transaction tax's burden on investors in different mutual funds?
Considering a sample of exchange-traded funds can give us clues to these answers. While daily portfolio holdings are not generally disclosed by funds, most ETFs are required to publish this data. For the purpose of this analysis, we looked at day-to-day changes in portfolio holdings over the entirety of 2019 for six ETFs. By calculating the daily difference, we could apply a tax to the market value of the shares sold or bought each day, replicating how the tax would be assessed. We considered a 0.10%, or 10 basis points, tax on all securities for this specific analysis as that has been the most commonly proposed structure. We took one passive and one active from each of three Morningstar Categories to test our suspicions that there would be differences in the impact to strategies based on whether they are active or passive and based on the market they target.
Even with a small sample size, initial findings track with these hypotheses. In all three categories, the impact to returns was greater for active funds than passive, ranging from differences of 0.013% to 0.081% in the reduction of returns between ETFs of the same category. This disparity is a result of the factors that trigger the purchase or sale of a portfolio holding. Flows, either in or out, constitute one trigger as the portfolio needs to expand or contract to maintain the correct weightings. Portfolio turnover, changing the holdings or the weightings of the holdings the portfolio is exposed to, would be the other main impetus for trades. In the case of indexed ETFs, the expectation is that most portfolio trading can be explained by flows as indexes only update every so often. Active funds would incur taxes from trading triggered by flows as well; however, they will additionally make trades based on strategic investment decisions. Cumulatively, as both a percentage of the portfolio and as a percentage of the return, this would result in active funds paying more financial transaction taxes.
Additionally, we observe that investments in a category with a stable group of holdings like U.S. large growth were generally less impacted than strategies investing in smaller markets or markets with more turnover, such as U.S. small-cap blend and foreign large blend. In this case, considering just the passive investments in each category allows our analysis to focus on the impact of the category and remove the element of how many strategic decisions a manager might make. The passive U.S. large-growth ETF saw only a 0.013% reduction in returns, while the U.S. small-cap blend was double that at 0.037% and the foreign large blend more than 4 times as much with a normalized 0.061% reduction. These differences once again make intuitive sense based on how frequently companies move in and out of the eligible markets for the indexes.
Initial Analysis Indicates Financial Transaction Tax Creates Specific Incentives The variation in a financial transaction tax's impact depending on investment strategy is interesting and could lead to an unintended consequence if it were implemented, such as further privileging passive vehicles, including in retirement accounts. Across the ETFs, we consistently saw that reduction in return led to large disparities in savings after 20 years, with some funds generating nearly 7 times as much in taxes. However, the absolute value of the total tax was not that high, bumping total returns down by a little less than 1 to less than a 1/10 of a basis point.
However, this is just a first analysis. As the policy research team undertakes the work of analyzing the full ETF universe, we will be able to draw clearer relationships between different factors and how much the tax impacts an investment. In particular, we will look at how the impact of this tax changes under different market conditions, recognizing that the examples here extrapolate the impact to portfolios in 2019--a good market year--over a 20-year period. We will also evaluate how the layering of taxation, through making regular investments rather than a single lump-sum purchase or through a fund-of-funds strategy, could dramatically increase total taxes. Both of these are relevant for 401(k)-savers given the paycheck deduction contribution mechanism and the common default investment of target-date funds. For now, the financial transaction tax is something for investors to keep on an eye on, while policymakers need to carefully consider whether these are the incentives they wish to create.