TECH TUESDAY is a weekly content series that covers all aspects of financial market technology. TECH TUESDAY is produced in collaboration with Nasdaq.
We’ve spent a lot of time talking about tick marks, stock splits, and perfect stock prices over the past few years. Indeed, research suggests that the right choice can increase stock valuations and reduce the cost of capital for companies. It’s good for investors and businesses.
But research also suggests that not all companies should have the same tick size. In fact, as stock prices rise, studies by academics, market makers, the United States, and even regulators in Europe have found that the best tick lets a stock trade with an included spread between 1 and 2 ticks.
Chart 1: Spread costs are lowest when stocks are trading 2 ticks wide
Today we look at more data from the tick driver that confirms this finding. We find the data created by tick pilot which proves:
- Constraining stock prices increases costs and hurts valuations.
- But too many ticks are also bad for trading.
- Setting tick sizes so that a stock trades over 2 ticks reduces costs and improves valuations, even if it means increasing tick sizes.
Constantly reducing tick size does not help all actions
The U.S. stock market has seen two major tick size reductions – first from one-eighth of a dollar (12.5 cents) to one-sixteenth (6.25 cents) in 1997, then to one cent following decimalization rules added in 2001. The data in Chart 2 shows how dispatch costs have fallen as a result.
This left all stocks with a 1 cent tick, where they (mostly) remained.
Since 2001, as stock prices have risen, the cost of that 1 cent tick (in basis points) has steadily fallen. However, spreads have remained stable and more recently there is a possibility that spreads could increase, especially for all high priced stocks in the market. A sign that too many ticks that generate too little economic profit to attract liquidity providers is also bad.
Chart 2: Decreasing tick sizes helped reduce average spreads noticeably, but since decimalisation spreads have actually widened in cents
We also know that decimalization has left many stocks trading with spreads over 25 ticks wide, even if we look at odd lot spreads. In short, for many stocks, the 1 cent tick mark is too small.
Chart 3: For many stocks, the 1 cent tick is too small
The Tick Size Pilot widened spreads on some stocks
In 2016, the SEC was convinced that too many ticks were a problem, at least for some stocks. And thus, the much-maligned tick pilot was born.
The tick pilot set ticks at 5 cents for a wide range of 1,200 (mostly) small cap stocks.
The problem with the tick pilot was that many stocks added to the pilot were already trading at or near their optimal spreads of 1-2 cents. And for these stocks, the tick driver forced their spreads much wider, making them “tick constrained” instead of having an optimal price.
The impact is clear in Chart 1 of this previous report – where tick leader stocks form in a row with much wider spreads than “normal” stocks but also longer queues (larger circles) . This is also clear in the data from the official Tick Pilot Report (left columns of Chart 4 below). It was also hoped that deeper queues would add to overall liquidity, but instead, liquidity in these stocks plummeted.
Chart 4: Changes in spreads caused by the new 5 cent tick
Given that investors typically cross spreads more than they capture, especially as liquidity has fallen, it’s no surprise that experts have suggested the tick pilot will cost investors between $300 and $900 million dollars over two years.
New research reveals that the Tick Size pilot also affected valuations
It turns out that these wider spreads didn’t just increase transaction costs. A new academic study finds that wider spreads also hurt the share price performance of pilot tick stocks. Specifically, he finds that the valuations of the newly constrained stocks fell by about 3% during the pilot (Chart 5).
Using a bottom-of-the-envelope calculation, the paper estimated that the price drop equates to a loss of approximately $2.6 billion in market capitalization.
Chart 5: Price impact for tick-limited stocks
We find that spreads and valuations improved for stocks when the tick was “right sized”
But it got us wondering (given how we think of stocks with “too many ticks”), what happened to tick pilot stocks where a 5 cent tick was closer to optimum for them?
We therefore turned our attention to determining whether stocks with spreads in the “10 cents or more” category – where a 5 cent tick suddenly widened their trading by 2 ticks – were closer to the “optimum” on the basis of all the other research we discussed above.
The data in Chart 4 shows that spreads for these stocks have actually declined by around 8%. So setting “too many ticks” can actually reduce spreads by encouraging liquidity providers to significantly improve spreads if they want queue priority.
More importantly, we found that stocks where the tick driver “adjusted” their tick (from too many ticks to a 2-tick spread), on average, actually outperformed.
Chart 6: Impact on stock prices dropping to 2 ticks
In short, increasing the tick size to 5 cents was actually good for some stocks, especially those whose natural spread was around 10 cents. This is consistent with the choice of a tick that puts the spreads on these stocks close to their optimal return for tradability – where the 5 cent tick caused them to trade 2 ticks wide (a natural spread of $0.10 = 2 x 5 hundred ticks).
As a result, these stocks not only saw their spread costs decline, but they also outperformed.
Important implications for future tick size design
The tick pilot was ridiculed because he unnecessarily added trading costs to so many stocks.
But this new research also shows that stocks can also have too many ticks – and that consolidating ticks when they do also lowers investors’ costs by tightening spreads, which in turn improves valuations and, by extension, therefore, the cost of capital for a business.
In summary, the results underscore the importance of getting the right tick size – for all stock prices – by breaking down spreads for tick-limited stocks as well as consolidating quotes for stocks with too many ticks according to the current rules.
It also shows that one tick size does NOT fit all stocks. Something regulators in many other regions have known for some time.
Phil Mackintosh is chief economist at Nasdaq.
Shiyun Song, senior economics and statistics researcher for Nasdaq Economic Research, contributed to this article.