False Economy

Several times, I've seen expensive development staff being slowed down by the inadequacies of their infrastructure -- both technical and organisational. My ideal in this respect was a brief spell at Digital's (now Compaq's) SystemsResearchCenter? where everything just worked. In many other places, the developers waste time jumping through institutional hoops to, say, get a network connection working because the support system has been designed by accountants or for generic clerical workers.

My metaphor for this is the humble London bus. We still have many buses with the entrance at the back, that require both a driver and a conductor. Most bus companies would like to move to one-person buses because the staff costs are lower, but there is an interesting side-effect to this conversion. One-person buses take more time, because the driver also has to collect fares at each stop, so they slow down the whole traffic stream. The saving for the bus company makes everyone else's life more difficult and expensive. --SteveFreeman

See also TragedyOfTheCommons. The Commons in this case is the development staff's time. Managers look for ways to simplify their processes. Often the result is really burden-shifting away from one person onto everyone else. The additional burden on each individual employee is deemed to have a negligible effect, while the cost-savings on the one person is considered substantial.

Let's distinguish between the temporal and the spatial. FalseEconomy results from excessively local concern. You want cheapness now even if that means paying more in the long run, or you want cheapness here even if that means paying more over there.

Interestingly, XP explicitly discounts the future so we should naively expect it to produce the first kind of false economy, the temporal kind. -- DaveHarris

I would say that FalseEconomy comes more from not working out your costs properly. It is perfectly reasonable to discount the future against current savings if you know that is what you're doing. Too often, the short-term cost savings are more to do with organisational boundaries (i.e. NotOnMyBudget?) than an understanding of the situation. There's a whole accounting thing called ActivityBasedCosting? which is supposed to address this. -sf

With Steve, I would say that FalseEconomy is due to a mismatch of perceived costs to real costs. For example, XP argues that much building for the future to save time/cost then is a FalseEconomy. We would also argue that building for now without refactoring is a FalseEconomy (even in the middle term) because it will slow you down too much.

There can be far-sighted and near-sighted false economies. Doubtless there are other kinds as well. Use of frameworks, pro or con, is a locus of potential false economies, for example. --RonJeffries

The best of intentions often lead to UnintendedConsequences. It is fun to look at legislation in the political domain as well as decision-making in large organizations... people believe that they are optimizing what they hope to optimize, but with what side-effects? It takes very stern thought to uncover UnintendedConsequences prior to making a decision. -- MichaelFeathers

FalseEconomy: driving faster so that you get to the gas station before the tank runs dry.

My theory is that bad metrics drive false economies in corporations.

Companies engage in strange false economies. I've seen highly paid programmers sitting on their hands for days waiting for a network connection because the company economized their (relatively) low-paid support staff -- they made sure there was just enough support staff to do the job, with no waste. There is no way to achieve this kind of economy without developing a queue of work; if there were no queue, or a very short queue, your support people would sometimes run out of work. Waste!

I've seen the false support economy at work in many places. My theory is that it results from upper management applying the wrong metric to the support manager. If "cost" and "efficiency" is the number one metric for that manager, what will he do? Yep. Make everyone else wait so that his numbers look better. What if the cost of people waiting for services were factored into that manager's metrics? "George, you saved $100,000 on support last year, but the cost of people waiting for service rose by $200,000. Not good."

Good point. The PointyHairedBoss from Dilbert said it best, along the lines of "they haven't made the metric yet that can defeat me!". The metric was the repeat customer rate; he listed 100%, but didn't do so well if you eliminated warranty exchanges... [also where the metric was employee turnover: They were too disheartened to leave]

The problem isn't that we use the wrong metrics, but the practice of worshiping the metric. If you give me a metric, nine times out of ten I can do something that makes the metric very happy while making the business very unhappy. You need people to realize when a metric is being counterproductive, or merely being "defeated". --RobMandeville

I've seen another false economy in travel services. Many large corporations require you to book all business travel through their travel services. Here in Phoenix, I can fly to LA on Southwest Airlines for $100 round trip, and stay in a decent hotel with a microwave for $60. GE required me to book travel through their travel services, which bought me a $180 ticket on a different airline and stuck me in a $120 tower (with no microwave!). It turns out that GE negotiates "discounts" with certain airlines and hotels, and in order to realize those discounts, they only purchase travel services from those places. If they don't book a certain percentage of their travel with those firms, they can lose the discount.

Question for managers: Why would a sane manager willingly give a monopoly contract to any supplier? Put yourself in the shoes of the airline or hotel. Are you going to enter into that contract so that you can make less money from the customer, or more?

My theory is that, again, the wrong metric is being applied to the manager for travel services: That the manager is rated not on how much is spent for travel, but on how much is "saved." The money that's spent doesn't come from travel services -- it comes from the traveler's office. So at the end of the year, travel services gets to show "look, we negotiated a 20% discount on 5M worth of travel... look how much money we saved!" Never mind that the local organization is paying more and getting worse service. If travel services were instead rated on total amount spent per passenger mile (for example) and dollars lost due to travel problems, a sane manager would do everything possible to optimize for that metric.

-- WayneConrad

Often in manufacturing settings, a production center is billed an overhead on their labor costs to pay for all the goodies that they consume, electricity, accounting services, etc. This is sometimes quite large, say 200% or more. Let's take 200% as an example. So a manager makes a product that has $1000 of labor, and is billed $2000 for overhead. Now he outsources, pays $2000 to bring the product in the door, and on paper has saved $1000. Typically the actual cost to the company for overhead much less than 200%. The manager has cost the company a lot of money while looking like a genius. A FalseEconomy. -- RobertField

This kind of problem affects everything, including politics because people discover that manipulating highly visible metrics is easier than satisfying a wider range of metrics. Thus, the "economy" becomes a perception game above a merit game, and the culture of perception tinkering rots the system as perception tinkerers grow in power and run the show. Merit-oriented workers are ignored, stuffed in a dusty corner, and kept uninformed.