The perils of financial predictions

I’m reading a 2011 book called Guaranteed to Fail: Fannie Mae, Freddie Mac and the Debacle of Mortgage Finance by Viral V. Acharya, Matthew Richardson, Stijn van Nieuwerburgh, and Lawrence J. White. This is a serious book by reputable scholars. I’ve made it to p. 3, where they write:

[T]he chances are slim to none that either Fannie or Freddie will be able to pay back the funds [they received from the government]…. So where is the outrage?

And yet as we know, Fannie and Freddie have paid back the government (or, more precisely, the government has made money on its investment in preferred stock).

Here is the paradox or irony or whatever you want to call it. During the financial crisis, we learned that the CDOs and other mortgage-backed securities that everyone thought were safe were based on assumptions about housing prices and underwriting standards that turned out to be wrong. Their market value plunged as people stopped trading them. The government saved the day by lending widely so people could hold onto these securities until maturity or reduce their exposure to them.

The proper inference was “these securities are hard to value.” The actual, wrong response was “these securities are worthless [or close to worthless].” Of course, if the securities are hard to value, we don’t know whether they are worthless or not. And yet it seems that nearly everyone draw exactly the wrong lesson. People had as much unwarranted confidence in their valuations of the securities after the crisis as they did before; they simply changed what they thought the valuations were–from high to low. And they were wrong again.