Kathy Sierra has a great post on the degree to which software designers should design in user freedom.
On the face of it, freedom is good. More freedom is better. In fact, if it doesn’t threaten us bodily harm, then more freedom is way better. Isn’t it?
Not so. Sierra offers a 2×2 matrix relating payoff to effort. The payoff is good for things like Tivo or Amazon. But digital home thermostats and new stereo systems give us too much freedom for the payoff. They’re just a pain.
There is a limit beyond which freedom of choice generates shutdown. Barry Schwartz’s The Paradox of Choice explores it well. After a while, complexity overwhelms the desirability of choice.
Sierra and Schwartz happen to illustrate the economic relationship between freedom and trust. In a nutshell, we give up freedom of choice in return for more efficient use of our time. We do it with trust.
Branding is the corporate version of trust. Rather than analyze every brand of bottled water, every version of jeans, or every make and model of HD-TV, we abdicate our freedom to do so in return for the security of a brand name. We trust Sony, or Coke, or Amazon, to make acceptably acceptable selections for us—so are freed to make other decisions.
But trust is about more than branding.
The last two centuries of global economic development have been driven by the search for division of labor. Adam Smith’s pin-makers organized around 19 specialized operations; it was far cheaper to assign individuals to distinct operations than to have each operator do all operations.
The transaction cost of coordination was well below the benefits of specialization.
At a corporate level, transaction costs remained high at the turn of the 20th century; early US auto companies made their own tires rather than incur the cost and risk of buying tires from others.
As transaction costs declined, it became more feasible to contract work out—the history of the auto industry is one of moving from integrated manufacturers to contract assemblers.
In recent years, we’ve seen diverging trends: lower unit transaction costs, and higher volumes of transaction costs. The net effect has been driven more by volume than by unit cost. Transaction costs as a percent of GDP have been going up. By one estimate, they now exceed 50% in the US.
We are reminded constantly of the internet’s effect on lowering unit transaction costs; but we don’t notice that the total of such costs is rising.
Here’s what it means: for further economic efficiency, the ability to reduce transaction costs is going to become more critical than further division of labor.
The more technically and globally integrated we get, the more freedom of choice we get. But at some point, freedom of choice becomes overwhelming.
If I want to make and sell jeans, I probably have dozens (hundreds? thousands?) of ways to contract the work out. Past some point, I don’t want more options—I want someone I can trust to make that decision for me.
In other words, I’ll give up freedom in return for lower transaction costs. The currency of that exchange is trust.
In an economy where half the costs are transaction costs, the currency of trust is massively valuable. Think of the transaction costs between auto producers and their suppliers: lawyers, agreements, contracts, specifications, bonus systems, QC, compliance, etc. Suppose they were 100% obliterated by trust. What kind of marketplace cost reduction would that provide?
Trust is not soft stuff. In a world that is getting massively more connected, greater trust has a very real economic role to play.
Giving up freedom for trust can be, paradoxically, a very freeing thing.