Is this as good as it gets? How much can social risk ever decline?

Karl Smith reports:

Are we so sure there is a better way?

Real interest have famously been on the decline for thirty years. A rougher historical record suggests that English real interest rates may have been in decline since at least 1600.

The standard explanation here is better governance and lower systemic risk.

Yet, lets imagine a simple model where we have two sources of risk. There is background you cannot avoid. And, there is personal risk that you create by through your own choices.

Policy makers have since Thomas Hobbes been attempting to drive down background risk. They have larger been successful. As a result our lives are getting more and more stable.

As that happens, however, folks are going to tend to take on more personal risk. There is a tradeoff between risk and reward. As you face less background risk, for which you not rewarded it makes sense to go for more personal risk for which you are rewarded.

When I take on more personal risk, however, it bleeds over slightly into everyone else’s background risk. People depend on me. If I take risks and lose so big that I debilitate myself then my family and my friends will surely suffer. But, so will my employer, my creditor and the businesses who count on me as a regular. When I go down, they go down.

So, putting it all back together and we come up with something of a risk floor, if you will.

Policy makers drive down systemic background risk. This makes everyone safer. In response, each individual takes on a little more personal risk and contributes slightly to the general background risk.

Eventually we will reach a point where policy makers have driven out so much systemic background risk that any marginal decrease systemic background risk will simply induce individuals to take on more personal risk until they raise the total risk level back up to where it was before.

Safer policy then has little net effect.

Said another way, attempts to prevent bubbles from forming will only make folks more complacent about bubbles. Eventually, a bubble will slip through the cracks. However, folks will deny it’s a bubble, because don’t you know, bubbles are a thing of the past. Even as it grows to massive proportions the smartest minds will argue that it only looks like a bubble. If it were a real bubble, surely the Fed would have popped it by know.

And, so it grows larger and larger and larger. When it pops the downdraft is so great that policy makers don’t have the tools to deal with it. Perhaps, in a technical sense they do. They could stand firm on an NGDP target or pass the mother of all stimulus bills.

However, emotionally they are at a loss. They have never seen anything like this and until recently thought it was impossible. Now, they are being asked to approve policies that no one has used since the dark ages, while in the middle of a crisis no living person understands.

This is a task few people have the nerve to handle. And, so they don’t handle it and the downdraft smashes the entire global economy to bits.

Such, is often the problem of “over-solving” the problems of the past.


So when we will hit the age of 0% real English rates, or should we ask Zeno's tortoise when it finally reaches its goal ahead of fleet Achilles?'s_paradoxes#Achilles_and_the_tortoise

Real interest rates are already negative.

'Real interest rates are already negative.'

Interesting. I don't follow the pound much, but let's see -

'The annual inflation rate was unchanged in March at 2.8 percent, the Office for National Statistics said. Core inflation, which excludes alcohol, tobacco, food and energy prices, accelerated to 2.4 percent from 2.3 percent in February. The Retail Prices Index, used as a basis for the inflation- linked bond market, was 3.3 percent, up from 3.2 percent.


The 10-year gilt yield rose three basis points, or 0.03 percentage point, to close at 1.73 percent after falling to 1.63 percent on April 8, the lowest since Sept. 5. The 1.75 percent bond due September 2022 declined 0.23, or 2.30 pounds per 1,000- pound face amount, to 100.165. '

Quite right - real interest rates, for debt bought right now, are negative (and likely to remain that way for the life of the debt). But then there is also a certain perspective of time involved. If one is holding a 30 year gilt from say, 2003, bought on June 7th with a rate of 4.475%, the real rate of return does not look to be zero, much less negative, at this point.

This is a silly meme that:

1) Assumes that every person can perfectly calibrate the amount of risk in his own life. That's simply not possible - too many risks, many of them almost impossible to accurately assess. If you want to make this concrete, ask yourself if it was more dangerous this week to be in Boston or in West, Texas. Then look at the casualty statistics and decide which one was really more dangerous and which one "felt" more dangerous. You could make an argument for either one, I think - which is my point.

2) Assumes that every person is happy with the total amount of risk in their lives and does not want to see it go down. I suppose in theory there could be such an asymptotic limit, but given the amount of suffering and anxiety in the world today, I don't think we have to worry about approaching it.

Evidence from workplace accidents, crime statistics, auto injuries and a bunch of other places all suggest that it continues to be possible to reduce risk both in societal and personal terms.

Evidence from personal savings rates points the other way.

Yes, but. A lot of my fellow baby boomers seemed to think "the future would take care of itself", i.e. that they could socialize their old age risk. This may not have been correct.

Similarly, large financial institutions socialized their risk, correctly figuring the government would step in.

That premise certainly seems to have been correct. Why would you think otherwise?

Dylan, personal savings in the U.S. is declining, but there are other countries where it is staying about the same or rising. Not a very clear signal.
Maybe the best measure would be life expectancy, as that encompasses health, wealth and risk tolerance.

Interesting thought and there is at least some evidence that people (esp young men) are actively looking for risk. Making cars safer increases riskier driving behaviour, rise of extreme sports.

However, I don't buy the bit that we are now worse at reacting to the current economic situation than societies of the past. Were they better at preventing bubbles or dealing with recessions? I would at least like to hear some examples.

One quibble, Extreme Sports are popular because its an athletic sport that rich white kids can still dominate. No snow or 1/2 acre skateboard ramps in the Dominican or Vallejo.

It seems that this argument makes more sense if you think of behavioral and social risks more than recessions. Taking the last 40-50 years as the really experimental period, the good years of the 1950s lulled people into a sense of security. There was the view that growth would continue unabated (serious scholars made arguments that once you had reached take-off growth, there was no turning back and now all you worried about was redistribution) and so people experimented with different lifestyles, etc. Traditional values and norms were being overturned. Divorce, crime, unrest, etc. were all going up. It seems to me the penchant for increased personal liberation and the tension that has made for bourgeois behavior (see Murray's recent book on the 2 Americas) as well as the fad for non-assimilationist diversity are the really risky experiments. Maybe great if it works out. But if not? What if groups form that are willing and able to carry out intense acts of terrorism in the US on a regular basis. Is the liberal state really up to dealing with it? We may well find out. After all, it's not as if large sections of "bad areas" in the US are already ungovernable.

Interesting but a bit of balderdash seems to me. If personal risk "bleeds into" background risk, then it becomes background risk. So have or have not we reduced background risk? Perhaps the invention of the limited liability corporation is to blame, as well as liberalizing bankruptcy laws to make risk taking more acceptable? Not to mention moral hazard via depository insurance?

"If personal risk “bleeds into” background risk, then it becomes background risk. So have or have not we reduced background risk?"
Well, isn't that the point? We reduced it a bit, but not as much as we thought we would because we liberated people to up their personal risk, which bled back in. So every time we try to reduce background risk there is the real question, as you put it, have we or have we not reduced background risk. Sometimes we have, but likely never by as much as we expected. Sometimes we may even have raised the background risk. This is the whole concept the article developed. Very neat bit of thinking it is too.

Neat, circular and untestable. But that's economics--or Freakonomics--for ya! :-)

There's a lot of fatalism in economics blogs, and I've never understood why, for instance, we're supposed to accept that there's some ineluctable sweeping-arc-of-history that must have produced present trends and that's unavoidable (or, in this case, avoidable only through a return to government barbarism, which is the same thing). I suppose it makes things feel more dramatic. I say, let's try again after we've put finance back in the bottle either through complete nationalization or massive equity requirements, and probably at some point a big tax regime internationalization. Let's see how things change then, and after that we can do the sweeping arc of history thing.

There is something to this line of thought, but also some vapor.

The something to it goes about like this: for evolutionary reasons, people in general and young men in particular seek various kinds of status - and some kinds of status, some very desireable kinds, are only achievable by prevailing in contests (with at least a risk of not winning) or prevailing over hazards. In the sphere of economics, "risk" and "try some worthwhile new thing" are locked together.

So yes, there is a floor of risk where humans are involved, and a floor of risk for any reasonably healthy economy.

Whether these are changed by reduction of systemic risk is a different issue....

Fertile women are most attracted to men who live the most dangerously. It is women who will decide how many high risk takers there will be in the next generations.

Wouldn't risk *lower* real interest rates, because people would want to substitute into bonds as a means of hedging?

And isn't expected economic growth a big determinant of real rates? If the economy grows, we'll be richer in the future, and we'll need the money less. So it gives a time value to money.

I'm pretty confused about how the "lowering of the risk" is supposed to explain lower interest rates over the last thirty years. It might explain lower rates of return on risky assets like the stock market, though.

Is this a statement of economics or philosophy?

A more drawn out version of Minsky's "stability is destabilizing."

More drawn out than that 3 word phrase, but less drawn out than Minsky's multiple works elaborating on the idea--he makes much the same point at much greater lengths in various books and articles.

So, from 1935-1975, the systemic risk was extremely high which limited the amount of personal risk taking.

Then as the conservative thinking on banking affected financial regulation, the systemic risk was reduced and personal risk increased creating the bubbles circa 1987,which then led Reagan-Bush to really seek to reduce systematic risk with more push to deregulate leading to more personal risk in the bubble of 1999, followed by efforts to really crank down the systemic risks with more deregulation in 2000 which drastically increased personal risk with the bubble of 2007??

And the reason the bubble of 1999 wasn't more severe was the Clinton financial team were actively intervening in the bank system to introduce systemic risk to reduce personal risk... Eg, the bailout of Mexico was to increase systemic risk, forcing the investment banks to bailout LTCM was a move to boost systematic risk and reduce personal risk taking.


Didn't each era result in either a larger social and/or military spending (both justified by reducing risk)? Didn't government make ever larger amounts of private debt backed by tax payers (Fannie, Freddie, etc) justified by reducing risk to lenders in order to get them to extend more credit? Wasn't every deregulation done to effectively expand credit (eg lowering lending standards for Freddie/Fannie) to reduce risks that individual consumers could not find loans while increasing risk of background systemic collapse?

This post is fascinating and, in some ways, related to Alex's post about psychic harms. Ex ante (or ex post?), when a person takes on too much risk, is this really self-regarding behavior, or does it impose an external cost on others, as suggested by Karl Smith? That is, are one's risk-level preferences a matter of one's individual autonomy, or are such risk-level preferences an appropriate matter for public regulation?

Sam Peltzman should be referenced here.

This reads like an equilibrium would emerge. Total risk stays the same, but its components change. Governance risk declines, but is replaced by the increased background component of personal risk. Real interest rates are low today only for perceived low/no-risk borrowers, e.g., sovereigns. And that is only happening because monetary policy is way too tight given today's risk tolerance levels.

"Yet, lets imagine a simple model where we have two sources of risk."


This is just another attempt to argue that systemic action by the government will have unintended consequences and cannot work.

It's too bad the same type of reasoning isn't applied to the principle of the freedom to own guns ... [will lead to their inappropriate use].

This is just another _devastatingly successful_ attempt to argue that systemic action by the government will have unintended consequences and cannot work.

You missed a word or two.

The same type of reasoning is applied to freedom to own guns, by the way. It goes like this. People are, in the first instance, free to own guns. If a government decides (for some reason, who could imagine what) to take 'systemic action' to remove that freedom, there will be unintended consequences and it will not work. (It will not have the results the government says it intended; it may have the results that a malevolent government really intends.)

You're *such* a badass.

"The standard explanation here is better governance and lower systemic risk."

The biggest driver for lower interest rates is probably longer life expectancy.

Longer life expectancy may generally be lowering the appetite for all types of risk. Two hundred years ago a 30 year old who did something physically dangerous was risking perhaps 20 years of life while now he would be risking fifty years of life. With longer life expectancy, the cost of risky behavior goes up.

Managing personal risk isn't that hard. You just need more paternalism.

Oh, I see interest rates are down today. Time to go get into a high-speed chase on the freeway!

In other words, this line of thinking is full of curious economist-type assumptions such as the average person having any awareness of background risk. Perhaps if we're actually talking about the finance sector that makes sense, but otherwise it's a non-starter.

One's background risk for being lazy has been pretty much eliminated over the last half century. One's background risk for doing unhealthy things has been massively reduced, at least in places where medicine is socialized. One's background risk for having kids without having a husband has been vacuumed out. One's personal risk for being lazy or doing unhealthy things or having kids without having a husband has not, meanwhile, changed very much. But one can, and, the historical shows, does, assume more of it as the background risk declines.

I don't get what is so hard to grasp about this. The average person doesn't have any awareness about background risk? Really? Background risk is probably the foremost thing the average person has an awareness about.

Not surprising that "risk homeostasis" applies to financial decisions. The interesting question is how well calibrated the intuitive risk assessment is on which people base their chosen risk level and how fast they integrate new structural changes.

Perhaps new financial/regulatory innovations that people in aggregate don't initially understand throw off the homeostatic adjustment leading to the assumption of more risk than they want.

"Oh, I see interest rates are down today. Time to go get into a high-speed chase on the freeway!"

Well, interest rate risk doesn't affect driving risk.

But if the risk to borrowing goes down, then people will borrow more. So interest rate goes down, borrowing increases.

And once we get self driving cars and the risk of crashing decreases, the speed and complexity at which they operate should increase. The more stable the AI, the faster the cars, and the more complex their driving patterns will become.

But in both situations there are constraints. The amount of resources in the system. The laws of physics. This is what determines the risk floor. And as long as the artificial system enforces the constraints of the natural system, that's the solution.

Without background risk, everybody will borrow more. But as long as we don't lend out what we can't afford to lose, then there is nothing wrong with that. If we build a society where everybody has food and shelter and a right to live peacefully, then borrowing and lending can be about other things, like status and control.

I have my doubts this is the case:

1. There is no clear delineation between background and personal risk.
2. If income is exogenous, this clearly contradicts the basic assumption that marginal propensity to save increases with income. (Higher incomes mean lower background risk).
3. The idea ignores the fact that people are risk-averse, that is uncertainty and risk have costs ipso facto. (I would rather get an unconditional $1,000 than an equal chance of loosing $500/gaining $2,000). This implies a personal risk ceiling.
4. Rich European countries have way lower background risk because of bloated social safety net programs. But they don't invest in stocks either. Germany has 16% direct ownership rates; America 50%. Americans can also be fired a lot easier and face much more uncertainty. Same thing with Switzerland.
5. Risk aggregated on the personal level isn't necessarily same as risk for society as a whole. Equity and education are risky (less saving/more borrowing) but pay for themselves, creating feedback loops of decreased risk/higher future savings. Only in the short-term does his idea hold. Equity is riskier for the investor, but safer for the borrower. Risk of doing business decreases as people take higher "personal risks".
6. Obviously, I'm assuming (like him) that personal risk pays off. Not stupid sprees in Las Vegas.

As I say in my much more detailed post (, "My theory isn’t nearly as clean, and I definitely haven’t mathematized it. It’s anecdotal. And it’s not scientific. But I seriously doubt it’s as simple as he claims!"

Hi everyone, it's my first visit at this website, and article is truly fruitful designed for me, keep up posting these articles or reviews.

So tell me, how do you feel about automated posters that use meaningless, generic sounding statements to trick filters so they can boost their search score? This activity seems to be a rampant example of the "free rider problem."

Prehaps I am missing a key point. But don't we want to encourage risk taking ? Is there not a point of view that says more risk taking produces more innovation? So if this point of view is right, wouldn't the increase in negative effects of risk taking be offset by the positive effects of risk taking ?

An interesting thought that seems to me to fall into an untestable assertion rather than something we can meaningfully use. It also doesn't seem to account for any of the actions that triggered the last recession, namely people manipulating the risk management tools to pass a somewhat risky investment off as a safe one, or really any of the ones before that.

The assumption that less background risk leads to more personal risk taking was unsupported. "Hanged for a farthing, hanged for a pence" thinking would suggest the opposite.

Whenever someone talks about reducing risk, rather than shifting or repackaging risk, I'm all ears but suspicious.

Smith writes: “Eventually we will reach a point where policy makers have driven out so much systemic background risk that any marginal decrease [in] systemic background risk will simply induce individuals to take on more personal risk until they raise the total risk level back up to where it was before. Safer policy then has little net effect.”

Accepting his (dubious) conceptual framework, there is *no* effect on the total amount of risk, but there is a decrease in the proportion of total risk that is background and an increase in the proportion that is personal. This is important because (he says) there is no payoff to increased background risk, but there is a payoff to increased personal risk. So the economy is more productive if the proportion of background risk is lowered, even if total risk remains constant.

Comments for this post are closed