A study of human relationships may offer some insight into this question. For example, it is a well-discussed irony that people who โluredโ their partner out of a previous relationship are (or pretent to be) shocked when they are later on left by their partner themselves. But what can you expect from a person who has already proven to be someone who had no problem of ending a prior relationship? Statistics proves this point, too: While โonlyโ 40-50% of all first marriages end in divorce, the percentage for second marriages to end in divorce is significantly higher at 60% (Source: Dr. William H. Doherty, Utah State University)
The annual โdivorce rateโ for checking accounts is roughly 8-9% (Source: J.D. Power and Associates). I have no data available about a break-down of the numbers for customers who have previously switched banks vs. those who switch for the first time. But I would expect to find that there is a pool of โpromiscuousโ bank clients who will happily take their banking relationship to wherever short-term benefits are the highest. Having paid such a customers money to become oneโs client, it will be very difficult to โmonetizeโ the relationship. As precisely this customer has a proven track-record of switching banks the moment the free lunch is taken away (and replaced by high-fee structures).
So how can this concept ever work?