#2: Unplug The Blender

I hope you’ve all had some good opportunities to practice getting naked the past couple of months. I hope no one’s been arrested in the process!  This month’s suggestion builds on where we left things last time.

Here’s the issue. Our industry is often criticized for being overly optimistic about what we can accomplish through our development programs. William Easterly, an NYU development economist, once offered a compelling metaphor to describe this naïve ambition in his article The Utopian Nightmare. He joked that where the private sector had the wisdom to separately develop Rogaine for men to grow hair, and Nair for women to get rid of it, the development community would have spent years trying to develop one comprehensive product that removed hair from women’s legs and planted it on men’s heads.

I think Easterly has a point. It definitely made me laugh.  But I don’t agree that the cause of our industry's problem is naïve ambition. My own experience suggests the issue is less about what we set out to achieve, and more about what happens to our plans once the “rubber hits the road”.  To illustrate the point I’ve constructed another hypothetical.

Once again, you are running a private sector development project. This time you’re in Southeast Asia, and you're about to kick off a new activity. Your host country has an abundant supply of bamboo with some really unique colors and grain patterns. Experts think the bamboo’s unique qualities could enable it to quickly establish a US $500 Million footprint in the fast-growing global market for eco-friendly furniture and building materials. The problem is traditional drying techniques cause the wood to become severely cracked and discolored, making it unsuitable for export markets.

Your project recruited an international wood-processing expert to come in and take a look at the problem. The expert did a bunch of field research and outlined the following strategy:

The solution to the cracking and discoloration problem is careful management of the drying environment (i.e. humidity, temperature & oxygen levels). Determining the proper combination of these variables will require a controlled atmosphere wood kiln and up to a year of R&D effort. You’ll need to minimize the number of things that could go wrong during the R&D phase by: 

1. Working in partnership with the local agriculture university.   
2. Establishing just one processing site.
3. Providing rigorous oversight & meticulous documentation of every trial sequence. 
4. Importing a high-quality controlled-atmosphere wood kiln. 
5. Engaging the private sector to commercialize the process after the R&D is completed

The strategy is clear, simple, efficient, and reasonable. You present it to your client. The client loves it and gives you a green light to move forward. In my experience, this is how most new development projects start.

And then this happens …

Your client calls to tell you that the Embassy has rejected your request to import a high-quality wood kiln. She explains that there is an agreement with the local government to encourage growth in light manufacturing. When the Embassy heard your project was working on an initiative involving wood kilns, it made a commitment to the government that it would support manufacturing the kilns locally.

So, what’s your response? The expert’s strategy was clear: Import a high quality controlled atmosphere wood kiln for the R&D phase, minimize the number of things that could go wrong. Is this the right time to risk working with an inexperienced company that’s never built a kiln? You’re going to rock some high-level boats if you stick to your strategy and push back on the decision. You could undermine your project if you don’t. In a situation like this, unless you’ve already mastered the practice of getting naked, you’ll probably just go with the flow and start looking around for a local manufacturer.  While a single decision like this may not drive the project off its rails, it can start a cascade that becomes harder and harder to stop. To illustrate the point, lets go back to the example.

You’ve identified a local manufacturer that you think can figure out how to build the kilns. After several weeks of back and forth, the company says it needs a minimum order of 10 kilns to make a profit. You’ve only got the budget for one. The deal gets stuck. A few days later you get a panicked call from your client. News of the impasse has reached the Embassy and they are furious. They’re furious because they’ve just now learned that your project was planning to give a kiln to the agricultural university and don’t understand how that will create jobs and increase exports in the current year. Your client tells you that everyone wants to see a new plan for 10 private-sector wood processing ventures. They want to know how many jobs and exports the plan will generate by the end of the year. And they want it all by COB tomorrow.

What’s your response now? Once again, the expert’s strategy was clear: Work in partnership with the local agriculture university for the R&D phase. Establish one pilot site. Provide rigorous oversight & meticulous documentation of every trial sequence. Engage the private sector once the R&D is complete. Minimize the number of things that can go wrong. Of course, if you didn’t try to push back the first time, what are the chances that you’ll do it now? In my experience, slim to none.

So you get to work building out a new plan. You don’t have much time to establish estimates for job creation and export growth, so you do a back of the envelope calculation. 10 companies ... about 50 employees per company ... each with US$ 2.5 Million of exports in the first year. On the last line of your plan you type: "We estimate that the program will create 500 new full time jobs, and lead to a US$ 25 Million in exports in the first year.”

Now skip ahead a few weeks. Your back of the envelope estimate of $25 million in exports has made a real impression. The entire country seems excited about the program now. The local chamber of commerce pitched in to help you recruit 10 industrialists willing to set up processing companies. A large multinational donor agreed to provide credit guarantees through a local bank. The local government offered tax concessions in exchange for an agreement to discount the product for tribal artisans in remote, volatile areas. And, the flurry of activity convinced your local manufacturer to start building the first batch of kilns - and order parts and materials for a second.

At this point it’s really starting to look like a win/win for everyone. From its very humble beginning our little R&D program has morphed into the country’s new model for doing private sector development… all before a single plant of bamboo has even been cut.

I call this tendency to let simple, focused activities become elaborate, multi-stakeholder solutions as "Development Smoothies.”  You know... start with a banana (technology R&D), then blend in a scoop of ice cream (strengthen local manufacturing), add a peach (create 500 new jobs) throw in a couple of strawberries ($25 Million increase in exports), add some coconut milk to loosen things up (SME credit enhancements) and then top it off with a nice little cherry (new economic opportunities that help stabilize remote, volatile areas.)

Here’s the problem. Development smoothies may be great going down, but they have a terrible aftertaste. To illustrate, let’s fast forward through an entire year of our hypothetical.

Everything continued to go really well for a while. Each company received its wood kiln on time. The loans came through. Staffs were hired, ribbons were cut, and people got to work. Unfortunately, none of the companies managed to figure out a drying process for the bamboo that solved the cracking and coloring problems. One by one, each of the new companies concluded it just wasn’t possible and threw in the towel. By the end of the first year all ten companies had suspended plant operations, laid off the new employees, and stopped making loan payments for the kilns. The bank that made all the loans had to write them off, and got dropped from the sponsoring donor’s credit guarantee program as a result. Regulators reacted by suspending the bank’s license - its case is still unresolved, pending further investigations. The manufacturing company, sparked by that initial flurry of kiln orders, decided to make a bet big on the production line. Once the orders stopped it was forced to liquidate assets to pay down the debt, and eventually went out of business. And those tribal artisans in the remote, volatile areas that were promised cheap bamboo? They are now more convinced than ever that all their government knows how to do is make promises that it never intends to keep.

In a last ditch effort you decide to bring the wood processing expert back out to see if she can help troubleshoot the problems... identify a solution. She visits what’s left of the ten plants and discovers that none of them maintained any records. She also tested each of the kilns and found they all had leaks, and the controls weren’t properly calibrated. When she gets back to your office the expert pulls out the original strategy document, flips to the last page, and tosses it on your desk. “You didn’t import a high-quality kiln, you didn’t work with the university, you didn’t establish a single pilot site, you didn’t provide rigorous oversight, you don’t have any documentation of the trial sequences, and you didn’t wait to engage the private sector until the R&D was completed. Why do you think I emphasized each of those points specifically… what did you expect would happen?”

When a development smoothie starts coming together, it can be very hard to stop. Once it’s whipped up you’re stuck with it. There’s no going back. In our example, those fateful early decisions cased the R&D program to spin out of control, and led an entire country to end up mistakenly concluding that its bamboo simply can’t be dried. The scars may be too deep for anyone to ever really digest what happened, or to realize what was supposed to have happened instead.

That’s why, if we want to get really serious about development results, we have to learn to “unplug the blender.” Unplugging the blender means resisting the temptation to score quick points when it compromises key parts of a strategy.  It means taking more time away from busy people’s schedules to explain to them what you’re doing and why so they don’t make bad decisions for you. And it means fixing problems before they become unmanageable.

I think Easterly’s suggestion that problems with development stem from naïve ambition reflect his academic point of view. When you spend enough time in the trenches you discover that program designs are typically fine... the real problem is lousy implementation. That’s a problem we can solve, and its a problem we have to solve. Because while I may disagree with Easterly’s attribution of the problem, I can’t argue with his conclusion. By promising so much, we really do end up making the problems worse.

The Bottom Line

The development community needs to find ways to discipline itself with the same kind of bottom line focus that enables the private sector to make smarter decisions.

Next Up ... Learn to Drive