The Economics of Tick Regimes
We’ve done a lot of research on how to make trading cheaper for investors and reduce costs of capital for our listing companies.
One way is to remove the inefficiencies caused by ticks. We recently covered some of the research showing what the optimal number of ticks in a spread is. The overwhelming conclusion, using data from stock splits, real spreads, data from other countries and even the tick pilot, is spreads are cheapest when spreads are 2-3 ticks wide.
Based on that knowledge – simple math tells us what the optimal tick size should be (a half to a third of the spread).
Why are small spreads important?
- We also know from prior studies that the buy-side benefits, as on balance, they cross more spreads than they capture.
- We also know that issuers benefit from lower costs of capital and more liquidity when spreads are lower.
Today we join the academic research on optimal ticks with the actual spreads in the market and compute the economic costs of “wrong ticked” stocks.
We see that the SEC proposal may actually cost investors more (despite tackling the tick-constrained problem). However, alternatives exist where frictions are almost halved, a saving for mutual funds alone of around $362 million a year.
Cheaper spreads reduce trading costs
We start with our knowledge that spreads form in a U-shape even after accounting for round lot costs. That results in a goldilocks zone where spreads are cheapest – and an “optimal” tick that results in the tightest possible spread. It seems to happen where spreads are 2-3 ticks wide.
We show this in the chart below where:
- Tick-constrained stocks (blue zone) spreads cost more because of the tick constraint.
- Optimally ticked stocks (yellow zone) have the lowest possible spreads.
- Too many tick stocks (grey zone) even after accounting for round lots, spreads cost more (yellow line).
Importantly, the U-shape is flat at the minimum – so the difference between 1.5 to 4 cents (consolidation of all research) and 2 to 3 cents (intersection of most research) being optimal is relatively small.
Chart 1: Spreads start to widen as soon as stocks have more than 4 ticks
The chart above is just looking at large-cap stocks.
Because spreads change a lot more due to liquidity than ticks, it’s important to see if the curve exists for all levels of liquidity. We can see that in Chart 2 below. Fitting regression lines to each “ticks” group, Chart 2 shows that:
- 1-3 cent wide stocks (yellow) have a consistently lower spread than
- tick constrained (blue) or
- too many tick (grey) and the more ticks (black), the wider the spreads.
Chart 2: Spreads widen as liquidity falls, but for all levels of liquidity, yellow dots (1-3 tick spreads) stock have the cheapest spreads
We also see that there are a few tick-constrained stocks with very low liquidity (blue dots on the left of the chart). Importantly, this excludes SPACs and ETFs, which trade very differently due to their easier-to-value underlying portfolios.
How the current 1-cent tick regime works
If we want to see what might happen to tick constrained stocks in the future, we can look at effective spreads now (in cents). This should show us where stocks will fall in the “new” un-tick-constrained regime. We see in Chart 3 below that there are stocks with:
- Optimal Ticks: Around 2,700 stocks are “well ticked,” using the broader 1.5-4 cent classification that the research supports (yellow bars).
- Very tick constrained: What we call “tick constrained” is clearly debatable, but there are over 1,000 stocks with an effective spread of less than 0.8-cent now (pink bars) that surely seem to qualify.
- Tick constrained: Almost 2,000 stocks have an effective spread from just under 1 cent up to 1.5 cents, which, based on the research, could benefit from more ticks (blue bars).
- Too many ticks: Over 4,000 stocks already trade more than 5 cents wide, with around 450 trading more than 30 cents wide, which would be 6 x 5-cent tick.
Chart 3: Counting tick constrained, well ticks and too many stocks in the current market
Importantly, because the colors in Chart 2 and Chart 3 are largely consistent, you can see in Chart 2 which kinds of stocks have too many (and too few) ticks.
What the SEC proposals do
Based on effective spread data used in Chart 3 and applying the new SEC tick group rules, we estimate which tick groups stocks across the U.S. market will be added to. The data suggests:
- Just a handful of mega-caps will qualify for the 1/10th cent tick group, including T, EPD and UBS as well as ETFs like BNDX and IEMG.
- The majority of small and mid-cap stocks will trade with sub-penny ticks.
- Over 1000 microcap stocks will trade with less than a 1-cent spread, despite very few thinly traded stocks in Chart 2 being tick-constrained.
Chart 4: What tick buckets stocks will fall into based on market cap and current effective spreads
If we switch back to the same color scheme as Charts 1-3, we can see that many more stocks will trade with “too many ticks,” even in the 1/5th and ½ cent tick groups (all the black columns).
That’s because, under the SEC proposal, almost no stocks will have less than 4 tick spreads. In fact, as soon as a stock’s spread falls enough to be below 4 x its current tick, it is promoted to the next smaller tick group, as we discussed in our summary of the SEC proposals.
Chart 5: Trading under the SEC’s proposed tick regime
Given the U-shape that spreads form, it is more likely that spreads get wider under this regime, which we consistently see happens when stocks have “too many ticks.” We should also expect to see more odd lots fill into the new unused limit prices, just as we see now for stocks with “too many ticks.”
Ironically, that might have an effect similar to dark-volume-caps in Europe for some stocks, where:
- First, a smaller tick makes spreads wider,
- Which makes the stock qualify for a larger tick,
- Which makes spreads tighter,
- Which qualifies for a smaller tick, creating a cycle of changes.
Looking for more optimal ticks
Clearly, the data would suggest that tick-constrained stocks need a smaller tick. There is already some consensus around a ½ cent tick – including from including MEMX, Citadel Securities, XTX and Cboe.
A ½ cent tick for tick-constrained stocks
Hypothetically, if we use the SEC’s 1.6-cent bucket cutoff to define a tick-constrained stock but give all those stocks a ½ cent tick, we actually see a lot more stocks trading 2-3 ticks wide (yellow on the left side of Chart 6), which is optimal.
Based on effective spreads now, it seems some stocks may still trade close to tick constrained with a 1/2 cent tick (blue) – but that may only affect around 100 stocks.
A 5-cent tick for too-wide-spread stocks
Knowing that more than 4-cent spreads also add unnecessarily to trading costs, we could also create a 5-cent tick for stocks that already have spreads more than 5 cents wide. That is also consistent with lessons from the tick pilot to “make no spreads wider” or do no harm.
We could choose 8 cents as the cutoff for a 5-cent tick:
- That results in more stocks trading with a 1-cent tick (more normal), but also spreads with too many ticks (black bars) in the 1-cent tick group.
- However, stocks with an 8-cent spread would be set to trade 1.6 ticks wide (1.6 x 5 cents = 8 cents), putting them at the low end of the optimal range, turning that group in the 5c group back to yellow bars.
- Also, an 8-cent cutoff leaves no stocks in the 5-cent group being tick constrained – avoiding some of the mistakes of the tick pilot (where ticks forced spreads wider).
The results show much more yellow in the tick distribution (Chart 6). Although, because around 300 stocks trade more than 40 cents wide now, there are still some stocks that would still have too many ticks (black area at the right).
Chart 6: A proposal that tries to maximize “well ticked” (yellow) stock trading
There are other ways to solve this problem, such as moving tick group cutoffs or tick sizes slightly. However, they generally involve a trade-off between more tick constrained, too many ticks (especially around the tick group cutoffs) or more tick size categories (like other countries in the world). In reality, the majority of economic benefits are achieved even if you move cut-off levels slightly one way or the other.
What is the cost-benefit of each approach?
Knowing that spreads form a U-shape and that the buy side typically cross spreads a net of around 20% of the time and trades a total of around $23 trillion each year, we can estimate how much “unnecessary trading costs” wider spreads create (shown by the arrows below).
Chart 7: Plotting the average spreads from tick groups Chart 2 at different levels of daily liquidity, we see the U-shape persists across the spectrum of stocks, resulting in artificially wide spreads for tick-constrained and too-many-tick stocks
Using the math we created for stock splits (see Chart 8 here) and the expected spread changes from the dots we plotted in Chart 2, we can multiply by the amount of institutional trading in each stock to estimate the costs of “mis-ticked stocks” in each regime above.
We find that the different approaches result in different unnecessary costs caused by artificially wide spreads and less competitive NBBOs:
- Chart 3: $843 million per year - The current “one size fits all” regime adds to additional institutional trading costs via tick constrained and too many tick stocks.
- Chart 5: almost $900 million per year - Although the SEC’s new proposal fixes for the tick-constrained stock problem, it actually seems to add more costs to mutual funds by consistently adding too many ticks to “well-ticked” stocks, resulting in less competitive NBBOs.
- Chart 6: $481 million per year - The alternate proposal, where we try to set ticks closer to traded spreads, adds to a saving for mutual fund investors of over $360 million per year, or 43% of the frictions mutual funds currently incur.
Chart 8: Estimated additional costs from stocks being “mis-ticked,” resulting in wider than necessary spreads for investors to cross
What does this really look like to traders?
If we look at stocks' spreads across stock price, now (Chart 9) – the tick-constrained stocks are obvious (blue dots). However, so too at the “too-many-tick” stocks (black dots).
The data also shows that for almost all stock prices (any vertical slice), there are stocks that are tick constrained, as well as stocks with too many ticks. This is one reason why price-based tick groups used around the world are also sub-optimal – as stocks like MSFT, QQQ and SPY, with a 1-3 cent tick now, should not be forced to trade (say) 5 cents wide just because they have high stock prices.
Chart 9: What U.S. stock trading looks like now, colored by ticks in each spread
If, instead, we set ticks based on how the market trades each stock (similar to, but less aggressively, than how the SEC is constructing tick groups), we do no harm and:
- Let all tick-constrained stocks trade with a smaller tick so tick-constrained spreads (blue dots) are able to fall to more natural levels (Chart 8).
- This also allows stocks that trade well in the 1-cent tick to remain with a 1-cent tick (including, for now, IVV and TSLA).
- Given that stocks that already trade more than 5 cents wide ensure we do no harm (force no spreads wider), it also reduces pennying and odd lots inside the NBBO, which affects too-many-ticks stocks, helping them trade better (and closer to a 2-4 tick spread).
Chart 10: Representation of optimal new tick groups
What does this all mean?
We’ve spent a lot of time researching how to make stocks trade better, and stock splits prove that larger ticks (in too many tick stocks) can result in smaller spreads which also helps increase liquidity.
Most importantly, smaller spreads reduce trading costs for mutual funds, arbitrageurs and maybe even retail investors. They also reduce the costs of capital for issuers.
In short, getting ticks right is important.
Other Topics
Markets US Markets
Phil Mackintosh
Nasdaq
Phil Mackintosh is Chief Economist and a Senior Vice President at Nasdaq. His team is responsible for a variety of projects and initiatives in the U.S. and Europe to improve market structure, encourage capital formation and enhance trading efficiency.
Read Phil's Bio