As of late I have been some finance and economics journal as part of my final project. I combed through many virtual reams of paper in search of a topic to base the research on. It’s been a somewhat lukewarm search. See, the problem (or challenge, if you want to be a bit positive) with being ‘just’ an undergraduate is that I have enough knowledge to grasp what the paper might be about, but not so much that I can really understand what is being talked about. It’s like groping in the dark only with your phone screen on instead of that LED flash on the back. But that’s that.
The thing I want to note here is actually quite simple: that the bid-ask spread of a security can be used as a proxy to its liquidity, with liquidity here being the ease that the security can be transacted.1
In practice, measuring the changes of liquidity between two events (or periods, etc) can be done like so:
Let denote the average bid-ask spread of the security before an event (like repurchases, etc) and denote the average spread during (or after) the event. Calculate the difference in spread.
If then liquidity is assumed to have been worsened, and if then liquidity is assumed to have been improved. Statistical test should be used to before making these conclusions.
Challenges in deploying such a test include the availability and uniformity of data, general comparability, and finding ways to isolate the effect of the event alone.
1 McNally and Smith, 2011. The Journal of Financial Research. A MICROSTRUCTURE ANALYSIS OF THE LIQUIDITY IMPACT OF OPEN MARKET REPURCHASES.