Stats › Global markets and exchanges › Dark trading caps
Capping dark trading moved about one per cent of volume into periodic auctions and did not change trading costs
| Venue type | FCA estimate | EU wide estimate | Significance |
|---|---|---|---|
| Periodic auctions | +1.19 points | +0.9 to +1.3 points | Significant in both |
| Large in scale block venues | Not separated | +0.65 to +1.37 points | Significant under both estimators |
| Continuous lit markets | +1.29 points | +0.9 to +1.2 points | 5% in the first, not significant under one estimator in the second |
| Systematic internalisers | +0.12 points | Positive | Not significant in the first; mixed in the second |
| Execution costs | 0.519, standard error 2.780 | Intervals within 0.7 basis points of zero | No significant effect in either |
The first column is a difference in differences on United Kingdom transaction level data around the suspension and the later lift. The second is a European study of 1,082 stocks, 614 of them suspended, using both a stacked difference in differences and a regression discontinuity. Both papers conclude that banning one venue type did not raise investor trading costs where close alternatives existed.
The first calculation found 727 instruments over the market wide cap. Two independent studies then measured where the displaced volume went and both found periodic auctions absorbed roughly one percentage point, with no significant effect on execution costs.
Members only
The rest of this page is for members
Below this point there are 4 sections and 4 named sources, roughly 450 words of it. Every figure carries the source it came from and the date the data is from.
You can keep browsing every statistic in the library for free. The intro and the summary are always open.