The New York Times rolled out a paywall in March 2011. Unlike most paywalls, the Times paywall is easy to bypass. The process is well documented all over the web. A Google search for “how to bypass the New York Times paywall,” in quotation marks, returns over 100,000 results.
If the paywall is easy to bypass, then why have a paywall at all? Because it effectively segments the market.  
The first segment consists of customers with a higher willingness to pay that aren’t aware of the paywall workarounds or enjoy accessing the Times without futzing around. Who are these readers? Readers with higher disposable incomes, older readers, and heavier readers.
The second segment consists of customers with a lower willingness to pay, who have identified the workarounds, and are willing to deal with the additional steps involved in bypassing the paywall. Who are these readers? Readers with lower disposable incomes, younger readers, and lighter readers.
The result of this segmentation is that the Times has been able to attract a significant number of online paying subscribers (segment 1), without decimating its overall readership numbers (segment 1 + 2). By maintaining its overall readership numbers, the Times has been able to preserve its online advertising revenue.
The Times implemented an innovative paywall, and is achieving remarkable results.
I’d appreciate hearing your thoughts in the discussion below.
 Since the product is nearly identical for both segments, aside from the steps involved in bypassing the paywall, you might consider this as an example of price discrimination.
 A similar argument about segmentation was suggested earlier by Gizmo on Hacker News.