It is well known in the business world that requiring a credit card up front increases the barriers to becoming a lead (i.e. a potential customer). This decreases the number of leads a company gets, but increases the conversion rate, since only people who are serious about signing up will actually pull out their credit/debit card to begin with.
Quoting this study, it was found that 2% of website visitors would become free trials when a credit card was required, but 10% would become free trials when no credit card was required. This means that a free trial with a credit card is rejecting up to 80% of the people that would otherwise sign up for a free trial. Further, it is shown that 50% of people that sign up with a credit card go on to become customers, while free trials without credit cards only results in a 15% conversion ratio.
However, that only tells part of the story. While a 50% conversion rate sounds impressive, you have to remember that that is 50% of 20% of the number of leads that are acquired without a credit card free trial. While this study only had a sample size of 100, it suggests that companies actually do a lot better without a credit card during the free trial.
The no-credit-card trials get five times as many leads, and even though the conversion ratio is much lower, 90 day customer retention is better, ultimately resulting in more long term, stable revenue for the companies that do so. In other words, companies that do studies on maximizing profit have to decide if credit cards up front is better or not.
No matter how simple or complicated it is to cancel the account, the point is that credit cards up front produce fewer leads but with a better chance of conversion, while no-credit-card trials produce lower-quality leads but in a greater quantity. This means that companies have to choose on either better marketing to attract more leads, or better lead engagement to increase conversion ratios. It's all pretty complicated, but it's driven by a bunch of marketing research to try and figure out how to maximize profit.
There's no sinister motive in most cases (e.g. making it harder to cancel), simply filtering out those leads that they will have to spend money to cultivate only to lose them anyways. Each company decides if it's more profitable to have fewer leads at a higher conversion rate (usually because cost of cultivation is high) versus having lots of leads with lower conversion rates (usually because cost of cultivation is low).
There's other links in that post above that are worth reading, and with a bit of research, you can see that the decision to require a credit card for a trial is largely driven by the cost of cultivating a lead; higher quality leads are necessary for services that can't afford to give away too much "product" (or service, bandwidth, etc) for free, but some services may find it better to eat those up-front costs for the benefit of higher customer retention.