Magic Bullets vs. Multifaceted Interventions for Economic Stimulus, Economic Development and Weight Loss

*image source for the silver bullets above*

In "Obesity Is Always and Everywhere an Insulin Phenomenon," and the podcast "Miles Kimball Explains to Tracy Alloway and Joe Weisenthal Why Losing Weight Is Like Defeating Inflation" I compared successful weight loss to defeating inflation. In this post I reverse the direction of the analogy and compare successful weight loss to providing what it would have taken to get out of the Great Recession, as I laid out in "America's Big Monetary Policy Mistake: How Negative Interest Rates Could Have Stopped the Great Recession in Its Tracks."

Sometimes there is an intervention so powerful that it can solve a particular type of problem regardless of the specific causes of the problem and with only minimal assistance from complementary tools. For example, when economic stimulus is called for, deep enough interest rate cuts can provide the needed stimulus. As the posts linked in "How and Why to Eliminate the Zero Lower Bound: A Reader’s Guide" explain, for a smart, independent central bank, there is no effective lower bound on interest rates. So interest rate cuts can be the magic bullet for providing economic stimulus. Like any type of economic stimulus, too much is inflationary, but whatever is the right amount of stimulus, interest rate cuts can provide it. My detailed argument for the essentially unlimited power of negative interest rates to provide stimulus can be found in:

Just as negative interest rates are the magic bullet for providing economic stimulus, for those who can tolerate it, fasting—periods of time with water but no food—is the magic bullet for weight loss. Like negative interest rates, many people rule fasting out of bounds as a weight-loss tool, and end up instead with a complex set of recommendations that don't work very well individually, and are only halfway effective even in combination. Instead of ruling negative interest rates out of bounds, people should concentrate on making them work as smoothly as possible when they are needed. Instead of ruling fasting out of bounds, people should concentrate on making fasting as easy and painless as possible.

The secret to making periods with no food as easy and painless as possible is to eat foods that raise one's insulin index relatively little when one does eat. On how to do that, see my post "Forget Calorie Counting; It's the Insulin Index, Stupid." If you do that, you will be surprised at how painless it is to go for, say, 20 hours without eating. It is still important to arrange things to make it easy not to think about food by staying away from food as much as possible and having distractions from food such as work or a good TV show, but if you eat food low on the insulin index when you do eat, you will feel a lot less bodily hunger during fasting than you expect.

One thing I don't know is how long a period of time you need to have eaten low on the insulin index before fasting will become easy. Keto folks say it takes some time for your body to get used to generating energy from food that is not as easily digestible as sugar and other carbohydrates that have an insulin kick. In particular, the idea is that eating enough dietary fat can train your body to be able to generate energy from its own fat. If your body is good at processing its own fat, then there is a limit to how hungry you will be while fasting: once your blood sugar and insulin levels get low enough, your body will start turning your own fat into energy. If you get close to running out of body fat, you would have a problem, but most of us aren't that close to running out of body fat.

Here is my own experience: for the last half of 2016, I tried eating only 600 or so calories every other day without otherwise modifying my diet. I managed to do it, but that was hellish. In early 2017, I went off sugar, potatoes, rice and bread for several months. After that I started skipping meals and found it was easy. Also, I notice that when I cheat a little and eat some easily digestible carbs or other food high on the insulin index, I get hungrier and the fasting is harder.

Unlike negative interest rates, there aren't many people disputing the idea that fasting will help you lose weight. Instead, they claim that fasting is dangerous or hellish. I discussed hellish above. What about dangerous? Here I keep going back to this: surely human beings must be physically designed to go extended periods without food. Otherwise, how did our ancestors survive to become our ancestors? It just wasn't that easy to always have plenty of food handy two hundred thousand or thirty thousand years ago.

One important thing about a magic bullet is that it can work even if the set of all the things that affect something are complex. Many other things besides fasting can affect weight loss and weight gain. For example, in "A Conversation with David Brazel on Obesity Research," I mention the idea that keeping it cold—say keeping the thermostat at 68 degrees Fahrenheit instead of 72 degrees Fahrenheit—might aid weight loss. But fasting is such a powerful weight loss tool that you might not need to know every detail such as that. So I end up disagreeing with Marc Bellemare's very interesting post "Is the Study of Obesity Like Development Economics?" You may remember Marc from his guest post here: "Marc F. Bellemare's Story: 'I'm Bad at Math'" I am grateful for his permission to mirror "Is the Study of Obesity Like Development Economics?" below. Here are Marc's words:

I received a new book titled The Obesity Code last week, written by Canadian nephrologist Jason Fung.

Over the last year, I had read a few things by Dr. Fung on his clinic’s blog, but one of the things he says about obesity in his book made me believe that the study of obesity has a lot in common with development economics. Specifically, on p. 216 of his book, Dr. Fung writes:

The multifactorial nature of obesity is the crucial missing link. There is no one single cause of obesity. Do calories cause obesity? Yes, partially. Do carbohydrates cause obesity? Yes, partially. Does fiber protect us from obesity? Yes, partially. Does insulin resistance cause obesity? Yes, partially. Does sugar cause obesity? Yes, partially. All these factors converge on several hormonal pathways that lead to weight gain, and insulin is the most important of these. Low carbohydrate diet reduce insulin. Low-calorie diets restrict all food in there for reduce insulin. Paleo and [low-carb, high-fat] diets reduce insulin. Cabbage-soup diets reduce insulin. Reduced-food-reward diets reduce insulin. …
Obesity is a multifactorial disease. What we need is a framework, a structure, a coherent theory to understand how all its factors fit together. Too often, our current model of obesity assumes that there is only one single true cause, and that all others are pretenders to the throne. Endless debates ensue. Too many calories cause obesity. No, too many carbohydrates. No, too much saturated fat. No, too much red meat. No, too much processed foods. No, too much high-fat dairy. No, too much wheat. No, too much sugar. No, too much highly palatable foods. No, too much eating out. It goes on and on. They are all partially correct. …
Without understanding the multifactorial nature of obesity–which is critical–we are doomed to an endless cycle of blame.

In his book, Dr. Fung proposes a multi-pronged strategy to fight obesity. It looks as though he knows what he is talking about, since he has apparently helped several patients return to a normal weight.

Having long toyed with the idea of developing a model of individual weight dynamics wherein an S-shaped fat accumulation curve leads to there being two equilibria, much as in the modified Solow growth model with poverty traps, what the above excerpt made me think of is the equally multifactorial nature of poverty. Indeed, one of the first things I teach students when I teach a development course is that persistent economic under-development–poverty, that is–is the result of multiple market failures. The corollary of that statement is that there are no silver bullets in development. So unless you tackle all the problems at once, you might make a small, temporary dent in poverty, but people will tend to slide right back to where they were–much like people who, say, cut their calories tend to lose weight in the short run, but tend to go right back up to were they were after a few months.

In practice, this means that in situations of persistent economic under-development, microfinance will not magically lift people out of poverty. Nor will an input subsidy program. Nor will a micro-insurance scheme. Nor will any single intervention not aimed at resolving multiple market failures at once. And lest you think this makes me an advocate of something like the Millennium Villages Project, it does not. At least not in its current form, which does not lend itself to rigorous impact evaluation.

(By the way, this is why I had a hard time with Why Nations Fail: Because the authors are some of the smartest people alive today, yet their rhetoric seems to suggest that economic under-development has a single cause: extractive institutions. The journal articles on which they base the book are a lot more nuanced, so I imagine a publisher asked them to be slightly more polemic in order to make the book make an argument that sells better.)

If it were so simple as simply providing people with micro-loans, poverty would be a thing of the past. Similarly, if it were simply a question of “eat less, exercise more,” obesity would be a thing of the past as well. But whenever someone is suggesting a simple solution to a persistent problem, it is a safe bet that you are probably being sold snake oil.

I don’t know if Dr. Fung is the originator of the claim that obesity is multifactorial. But my intuition is that the claim is correct, and I certainly hope it gains traction with the publication of Dr. Fung’s book. As to what Dr. Fung advocates to treat obesity, you will have no doubt guessed that he does not advocate any single thing. Rather, The Obesity Code advocates doing all of the following:

Reducing one’s consumption of added sugars
Reducing one’s consumption of refined grains
Consuming a moderate amount of protein
Increasing one’s consumption of natural fats
Fasting for periods of 24, 36, even 48 hours
Getting seven to nine hours of sleep every night
Meditating

Of this list of things Jason Fung recommends, I argue that fasting is the centerpiece. The other things will help. And the low consumption of added sugars, refined grains and moderate consumption of protein is a key to making the fasting experience reasonably pleasant. But fasting is the big one.

Afterword: Returning briefly to economics, I really like what Marc says about economic development needing many interventions at once. I think that is right. Economic development can only be done in complex ways. Economic development is like getting healthy—something that requires many more elements than just weight loss. By contrast, economic stimulus, like weight loss, is best kept simple. There are many monetary policy papers talking about how to deal with the lower bound on interest rates, many advising things that are otherwise bad ideas. It is simpler and better to just eliminate the lower bound on interest rates! (See for example "The effective lower bound and the desirability of gradual interest rate adjustments," Sebastian Schmidt and compare this to my argument in "Next Generation Monetary Policy.")

Don't miss these other posts on diet and health and on fighting obesity:

Also see the last section of "Five Books That Have Changed My Life."