Success is knowing what worked yesterday.  The solutions that appear to work are often clear only in hindsight, and we humans can find a reason out of just about any set of circumstances. Real understanding is indicated more by the ability to predict, than the ability to explain why after the fact.

But even if we were correct about the cause of the outcome, it does not create a solution for all times and places. What worked before may not work today. The variables controlling the outcome, especially in economic and political frontiers, have likely changed so substantially that the same inputs may yield very different out outcomes.

Similarly what works for one size organization does not necessarily work for all size organizations.  Small businesses can afford to make intuitive decisions that may seems reckless to a larger organization. A small business can be far more flexible in responding to changing criteria, and feedback goes directly to the top manager. In larger organizations top executives are more often sequestered from such feedback.

Even among similar sized organizations there may be a stark difference in culture and structure that precludes the ability to share common solutions.  A company with significant debt may be less robust against sudden shocks that a debt free company could easily absorb.  The last financial shock even put companies with conservatively modest debt at risk. What had been deemed reasonable debt only months earlier jeopardized firms when the credit markets froze.

Models used to quantify risks gave delusional certainty that created comfort to levels of debt and risk  that proved disastrous. Models by their very nature exclude some variables to focus on others.  In the case of the financial system it was the ignored variables or at least the underestimated variables that proved fatal.  It was not enough to discern that the chance of loss was only 1% without understanding the size of the loss. In the case of our financial system the size of the loss was more than everything.

Part of the reason it seems so difficult to learn from the past is that we do not know what we do not know.  In matters such as economic policy we defer to academics who are reluctant to admit this vital shortcoming.  When they do fail they will insist that their model is still good if not perfect as if the odds of failure were so extreme that they should not be held accountable. Academics are reluctant to confess that a lifetime of work has been rendered worthless by a short bout with reality.

But a more likely reason we repeat mistakes is the preference for ideology over empirical evidence. We are all prone to hear what we expect to hear and see what we want to see.  Once a bad theory gets into one’s head, either in government or academia, it is hard to shake loose.