Hacker News new | past | comments | ask | show | jobs | submit login
Economists don’t understand much about the macroeconomy (qz.com)
55 points by elorant on Sept 19, 2019 | hide | past | favorite | 59 comments

Having worked at a successful macro hedge fund for more than 10 years, I can tell that:

1. There is no point in trying to predict macro economy further than 3 months

2. It's hard to predict which way the economy will go, but it's easier to foresee that there will be volatility or not.

3. If you want to make macro bets, you'd better have quant strategies that work in the case your macro bet goes South.

4. It's important to read and understand macro economists, not as much because they know what is going on, but rather because they are influencers, who participate in risking / de-risking movements

5. The most important things to understand in macro economics is that it is very much a people's problem. If you could get in the head of every central bank people, you'd get a strong edge over the market.

> Having worked at a successful macro hedge fund for more than 10 years

Do you mind sharing which one? (Ok, stupid question) Or whether it's a "quant" or global macro fund?

> There is no point in trying to predict macro economy further than 3 months

I disagree. Sometimes you can identify systematic factors. I remember Bob Litterman once telling me at a conference that a lot of their "quant" success was actually due to factoring in certain central bank policies that the central banks were willing to accept a certain, let's call it, slippage for in order to achieve certain goals. That apparently worked for years. Another example that comes to mind is the Euro convergence trade. Didn't that also go on for years? Unless you mean specifically GDP. Then yeah, I agree. In fact, I remember a class in financial econometrics where the teacher drew fancy equations on the blackboard only to conclude: "But you know, in practice a simple AR-model will perform just as well. It's just so you've seen it" :D

> It's important to read and understand macro economists, not as much because they know what is going on, but rather because they are influencers, who participate in risking / de-risking movements

Obligatory quote here:

"""You might suppose that the feud could be settled by testing ri- val claims. Alas, macroeconomics rarely works this way. Macro- economists cannot run experiments as laboratory scientists can. Statistical analysis of the world is muddied by the vast number of variables, many of which are correlated with the thing whose ef- fect the economist is trying to isolate. _Macroeconomic arguments tend not to produce winners and losers: only those with more in- fluence and those with less._ Post-Keynesian ideas were never pro- ven false, unlike the Ptolemaic model of the solar system. Rather, they declined in status as mainstream Keynesianism rose."""

[1] The Economist. Free exchange: Magic or logic? page 69, Mar 16 2019.

Which of course leads directly into your

> The most important things to understand in macro economics is that it is very much a people's problem. If you could get in the head of every central bank people, you'd get a strong edge over the market.

I've been saying this for YEARS, mostly to deaf ears. Thank you for confirming my personal biases/opinions ;) Finally someone else says it. It's why I (used to) like reading biographies, however shitty, of key people in power. Just to get a sense of their thinking or psychological makeup.

Very good comment, I'll keep it.

I would argue that macro-economy is done with the wrong timespan. Much climate and weather forecast, there is a gap that is very hard, if not impossible to fill.

Predicting the next semester is done quite alright. There are good economic models for long duration trend (50 years at least), though they are not used much because they are not very useful to invest as they do not have enough granularity. But the middle ground is just chaos.

I don't see anything in these statements about the usefulness of academic macroeconomics, except insofar as certain actors believe it. Is that right?

Scott Sumner (an economist) responded, I thought his take was spot on


Two quotes

> “Macrophysics” (predicting complex physical systems such as hurricanes and earthquakes) does just as poor a job of prediction as macroeconomics.

> The real problem is modeling the policymakers. We don’t know exactly how the 12 members of the FOMC will behave.

Just like chess. One can study openings, end-games as deep as possible, try to grasp the dynamics of positions but there is always the human element and that's very hard to study.

> “Macrophysics” (predicting complex physical systems such as hurricanes and earthquakes) does just as poor a job of prediction as macroeconomics.

Which is why people do not trust the weather men and neither should they trust economists.

> The real problem is modeling the policymakers. We don’t know exactly how the 12 members of the FOMC will behave.

This sounds hard. What about 327 million consumers? The real problem is a discipline in the social sciences which mistakes itself for hard science. It is serving today the same purpose "scientific socialism" once served in the Soviet Union: Taking power away from the people and placing it into the hands of an enlightended elite. What's called "depoliticisation" today. In which you remove spheres of public interest from the democratic decision-making process and cede them to unaccountable power. The exact thing we just haven't had enough of just yet according to this guy.

Listening to informed opinions is important but economists enjoy a kind of widespread unquestioned respect that has no basis in reality at all.

> Which is why people do not trust the weather men and neither should they trust economists.

People who's livelihood depend on the weather (farmers) do trust weathermen. They just know the limitations.

1: any forecast more than three days out cannot be counted on. (If you're close to mountains or other similarly confounding factor, that 3 day number is a lot smaller)

2: An 80% forecast of rain doesn't mean that it will rain in one specific place, but that it will rain in a good portion of the forecast area.

The same thing applies to economists. Some of their predictions are very accurate, but much are wishes and guesses. But due to human nature we primarily remember the times their guesses were wrong.

> This sounds hard. What about 327 million consumers?

Predicting what a lot of people will do is a lot easier than predicting what one person will do (which was the point of what you were quoting.)

We don’t know exactly how the 12 members of the FOMC will behave.

Oh, if only dim policymakers behaved as economists tell them to! Wouldn't the world be a wonderful place then?

Sumner agrees with your sarcasm, so not sure who you're arguing with ...

What he's pointing out is there's a human-driven force to economics so it's not accurate to compare it to a (theoretically) closed system like physics.

Macroeconomics is study of object with gradually changing dynamics.

Economists started to understand the dynamics of post-war type growing economy and how it behaves better and better. There is gradual shift away from these conditions.

* Wage share in OECD countries peaked late 70's early 80's and has started to decline.

* The return from increased education is slowing. Marginal return from increasing education is smaller.

* Women are now fully in workforce.

* Demographic change in OECD countries when people age. The share of working-age population is in decline. Paying for the care of older people dominates the growth of public spending and requires increases in taxation.

Now macroeconomists are trying to learn the dynamics of new conditions.

What all this amounts to is that there is no theoretical understanding of the the social interactions that constitute the economy, only fitted relationships to the actions of aggregating entities like companies.

That would require relatively complete "Theory of human" and also understanding the variety of human behavior. Human are also chancing their behavior. Different generations respond differently to same signal.

But all this does not mean that macroeconomics is useless or that it's not getting results. They are just relatively modest.

Well, he's not wrong.

It's a system which is slow to change, nigh on impossible to experiment with, we only have one experimental unit, and it responds to interventions. No wonder it's difficult to science it properly.

'Macroeconomics' by Williamson is my go-to example of the worst textbook I ever had to read (setting aside postmodernism). Overly certain, badly argued, with fishy math. So, a good place to go to for everything wrong with the discipline.

I think it was Tim Harford (possibly quoting someone else) who said "Microeconomics is everything we think we understand, macroeconomics is everything we don't".

My problem with macro-economics/economists is that so many of them seem so convinced aboout their own models and beliefs, or at least the conversation is dominated by those who are. It doesn't matter which view you have, there will be promonant economists who are convinced that it is the 'one true way" and the alternatives are bunk. Government policy then seems to swing depending on who is being listented to at the time.

I wouldn't mind macro-economists so much if they would just admit that a lot of thier views on the economy are largely faith, backed by cherry-picked data.

Does the opinion of economists even matter in practice even if they end up being correct? Politicians pass a lot of laws that simply do not make sense economically in the long run.

Here is an example: Globalization leads to concentration of economic opportunity in certain locations. If you simply go by the book then workers simply move to where the economic opportunity is and therefore everyone is better off than before globalization.

But local politicians of those locations want to avoid displacing their existing residents. Therefore they pass laws that make sense in the short term for their residents. Over the long term however they prevent the mobilization of all those future workers and the vast majority are worse off as a result of globalization.

When politicians create and force change, as a matter of dogma, and become known for carrying their books or following unproven theory from Hayek or Freedman, yes it probably matters. When those same politicians or economists point at the Chile miracle (that was a disaster), or economists become widely known as the "Berkeley Mafia", yes it probably matters.

Globalisation was about externalising costs and impacts. To hell with the local consequences - we got a decade or two of deindustrialisation, with lives and regions ruined for generations. Political fashions tend to stem from economic fashions without regard for the bigger picture. Or environment, citizens and residents.

Economics puts huge focus on material development, less so on health, culture or well-being. Your example is actually quite illustrating. When you do not have local politicians trying their best to keep their region relevant and populated, only those people who cannot leave for better opportunities remain. Within a city this means having ghettos, within a country it means having deep divisions. Globally it means you have a rich centre and a poor periphery. This might maximize productivity in a capitalist society, its certainly does not maximize happiness, and it exacerbates other problems like urban decay, crime, divisive politics and much more.

Economic as a discipline has a painfully narrow focus on mathematical modelling. Anything that doesn't fit into the models tends go get ignored: one of the reasons few economists spotted the global financial crisis is that their models didn't allow for such a crisis.

It would be silly to expect a biologist to write down an all-encompassing equation describing a cat, solve it, and thereby understand the animal. Why should that approach work on something infinitely more complex, like a whole economy?

I think that part of the problem is that macroeconomists think of the economy as a single system. Therefore, if you change a financial parameter, such as short interest rates, it will change supply and demand in the business economy.

It seems that the financial economy, which is the trading of various financial instruments between a lot of different parties, and the business economy, which produces, distributes and consumes goods and services, are two loosely coupled systems. Monetary volumes of financial trading are one or two orders of magnitude greater than the monetary volumes of good and services. Further, the financial economy is even more globalized than the business economy. The financial economy is showing signs of instability as regulators attempt to use it to control the business economy. Direct control of the business economy by fiscal measures, subsidies, tariffs and regulations would be more effective and cause less distortions than continued use of financial measures.

I've always thought that macroeconomics should be studied with a realistic perspective, and that does somewhat seem to be the direction economists are going with agent based modeling. Though perhaps too much on the math side and not enough fundamental thinking.

This is how I've always done my analysis. Start with 1000 people. Group them into 500 households along an age curve. Give each person a money account and assets along a wealth curve. Set up 100 businesses and the government. Study the dynamic system and gradually increase the realism/complexity with central banking, credit cards, etc. This is the hard unsimplified way but you will uncover the absolute truth of our system.

weird/crazy idea: using individual transaction data from various sources (credit cards, cash deposits, etc), would it be possible to build up a macro view from 100s of millions of people?

Start small (not the US), in a geo where electronic payments dominate, and where privacy isn't as important. If the model proves valuable, build modeling tools and scale up.

Counterpoint: there's a very large percentage (by value) of transactions that are invisible e.g. business partnerships, cryptocurrency, etc.

agent based modeling where the probability of the agents' actions is dependent on actual economic data...

more realistic sim city (city skylines)


I don't know that they've managed to produce the sort of general results that older economic theory.

> [Look at] the role of the bond rating agencies in 2008, which is completely unreformed. Why would that be? Well, there must be a lack of consensus.

I think we have a an obvious reason for the lack of consensus here, which is cognitive/regulatory capture. You'll always find someone willing to muddy the waters for a little cash and some sort of career advancement. Economics is the ideological science _par excellence_.


>Gates is right, in a way: Economists don’t understand much about the macroeconomy. No one does. Any responsible economist is the first to admit that.


> Economists’ research has contributed to fewer people living in poverty, low predictable inflation, and less risk and uncertainty.

How exactly?

Because macroeconomy is a large subset of economy, and wereas some subset of macroeconomy are well-understood, how those subsets interact with each other and with human/group psychology, however, is way harder to understand.

But also, you can be right for the wrong reason, especially in "soft" sciences.

The article presents that economic research was causal in these wins (poverty etc). There is no justification given for that in the article, it is pure dogmatic belief.

Said another way, how do we know these macroeconomic improvements are not in-spite-of this research?

Are you calling for a debate about the nature of cause and effect, or do you have a specific observation? Do you have some evidence that completely unmanaged economies have performed similarly in those measures?

As the GP states, the article accepts the claim:

"Economists don’t understand much about the macroeconomy"

But then says that in fact economists have e.g. contributed to fewer people living in poverty. I'm simply saying I don't accept that the research of economists has improved e.g the poverty situation without some evidence of that fact.

Before the discipline of 'economist' existed human societies experienced rises and falls of poverty, inflation etc. Is that sufficient evidence for you that 'completely unmanaged' (your words, which I don't even really understand your point here -- are you claiming economists 'manage' the economy?) economies have in the past had periods of ups like ours today?

The article is bordering on claiming that economists have nfi what they are doing, but the effects of their efforts have been amazing.

Does it make sense then to limit the impact by de-globalising economic activity to a scale that can be more easily understood?

How do you do that without also limiting the up-side? Globalisation has lifted hundreds of millions of people out of poverty. Localising economic activity would undo much of that.

Since 1970 the world has added about 4 billion people, I don't think the claim that "100 million" have been lifted out of poverty can be taken by itself. What if 100 million have been added to the "not super poor" class while 3.9 billion have been added to the "effectively" slaves class? Not saying that's what has happened but "lifted hundreds of millions of people out of poverty" sounds like a focused view that ignores all else.

I didn't say 100 million, I said hundreds of millions. That's a very cautious low ball though. That's probably the rough scale of people added to the middle class, not just raised out of poverty, in China alone. We don't need to guess what happened to the global population though, they're out there in the world. We can ask them.

Just since 1990, the share of the global population living on less than $1.90 a day has fallen from 44% to about 9.5% in 2015[1]. The graphs there are broken because different studies have used different measures of poverty at different times, but the trends and absolute numbers are clear. Global poverty as a proportion of population is about 1/6th what it was in 1960.


According to official statistics, China's middle class grew from 2.1% (29 million people) in 1999 to 39% (531 million people) in 2013. Average wagegrowth has been 11% between 2001 and 2015.


Excellent, that paints a much clearer picture.

> in China alone.

More like in China almost entirely. Global capitalist structures take a lot of credit for improving living standards all over the world, but when you look at the numbers, it's really a very dramatic improvement in China, a country that deviates from those structures in significant ways (and is demonized for ir.)

You should check out the book Factfulness by Hans Rosling. Rosling, a now deceased Swedish doctor, makes the case that the world has improved immensely over the past two centuries. He backs it up with copious statistics from the UN and other international NPOs. Rosling also gave one of the best TED talks ever: https://www.youtube.com/watch?v=hVimVzgtD6w

globalization only started in the 80's

Check out the book... included in the coverage of the past 200 years is coverage of the last 40.

The cost of that is considerable. No more cheap smartphones means no more app economy. If you want to go even further, the only way to really de-globalise is to stop using oil ...

OTOH no more gig economy shenanigans, VC dollars destroying industries, less of a race to the bottom everywhere, less centralization, probably healthier competition...

I don't have anything close to a belief yet, but the more I hang on HN and read stories from various industries, the more I'm starting to feel that maybe we took globalization a bit too far, or at least we're now learning about bad side effects. Human groups tend to be recursive hierarchies, and to this environment our intuitions about trust and behavior are adapted. Global economy does its best to cut straight through that.

Many smart people must have already talked through this to death. If anyone has pointers towards reading material, I'd strongly appreciate it.

Globalisation is fundamentally about individual freedom. It's about removing the barriers between a person in the USA or Europe meeting and doing business with a person in India, China or Africa without governments getting in the way. Fewer trade barriers, fewer regulations, more freedom.

When you give people freedoms like that, and there is huge advantage to be gained from it, people quickly find ways to scale it up to benefit the maximum number of people as much as possible. That's what a company is. It's a way to scale up economic activity.

This dynamic has raised hundreds of millions of people out of poverty in the last few decades alone. That's a number so big and a transformation in people's lives to dramatic that it's hard to grasp mentally in a meaningful way. That didn't happen because of farmer's markets and buying local produce in Wholefoods, much as it might pain us to admit it.

Does globalisation have it's down side? Yes, absolutely. It has increased pollution and accelerated the rate of consumption of resources. But what could we say to those families, those hundreds of millions of people, that could justify denying them their opportunities?

Fair. But on the flipside, what could we say to those families, those many millions, whose jobs and local economies get destroyed as multinational chains open shop and start arbitraging labor costs? What to say about rapid decline in quality of every good on the market, and the inability to even find a quality good if you have money? There were trust relationships that sort-of regulated this, but now I can't even find a decent furniture source, as both low and high-end brands ultimately have the same cheapest garbage parts somewhere in their supply chain. Consumer laws fix the worst problems here, but it's not perfect.

I'm not saying we should revert globalization; I'm right now only trying to learn about the good and bad sides, their relative balance, to figure out whether and where something needs to be done to achieve a better outcome for everyone. Because I get this growing feeling that screwing up local communities in exchange for freer money flow isn't turning as good a deal as advertised.

Your initial paragraphs talk about freedom, and I appreciate this view coming from a principle. Though looking at the world as it is today, couldn't you say that this focus on individual freedom, scaled up, actually ends up taking freedom from individuals away? How come when talking about freedom we focus on freedom to choose, but ignore the increasing constraints, the disappearance of options to choose from?

You describe capital freedom though, it's certainly not the case that immigration, that is actual, individual freedom, has been what raised most of those people out of poverty.

You have the problem exactly upside down. People resort to migration in order to obtain freedoms not available to them where they are. Only someone living in a cosy advanced free society, oblivious to the extremely constrained situation of the actual poor, could make such a profound error.

Billions of people worldwide didn't rise out of poverty since the 60s from migration though.

Capital belongs to people. Capital freedom is people exercising their rights of ownership over their property. That's what capitalism is. I know this very well because my wife is Chinese. Her father's family were villagers in the Chinese countryside, but they owned a few plots of land they rented to other farmers otherwise like themselves. As a result they were branded 'class enemies' and persecuted ruthlessly. To this day he sometimes tears up when we talk about those times, although I speak very little Chinese myself.

Globalisation is a term describing the deregulation of goods and capital, it is not at all related to migration.

The millions of people that have been lifted out of poverty in India and China are obviously still there, they haven't migrated. Globalisation is said to have helped people out of poverty by the increased flow of capital it represents, not by literally having them all move to richer countries.

In the face of the daily deaths along the "illegal" migrant routes, the idea that globalisation has made personal movement easier is unfounded.

I don't know why you keep talking about migration. It's not really relevant, in the grand scheme of things it's a rounding error.

Corporate freedom, not individual freedom. Make your shite without government getting in the way with pesky regulations about adequate safety, working conditions, environmental impact and a living wage, etc. Move to somewhere else if you can make it a dollar cheaper. We're already seeing places lined up to replace China for cheapest production. Those opportunities might turn out to have been fleeting.

Individual freedom. The freedom of entrepreneurs in China to build factories and make deals with the West. The freedom for Chinese workers to apply for and take jobs in those factories. They're all individuals. You can't take those freedoms away without hurting individuals either. We don't pass laws that apply on a company by company basis. A law that allows one entrepreneur to start a new business can allow a thousand others to do the same. That's how capitalism works. It's what it is.

China and the USA didn't pass a Jack Ma law that allowed Jack Ma to trade with America. He took advantage of individual freedoms available to anyone. I'm sure he had particular opportunities too, but he was free to take advantage of them, as others were free to take advantage of opportunities they had.

China is working it's way up the economic ladder. Low end assembly isn't moving out of China because other countries are getting cheaper. It's moving out because there are more higher paid jobs in China now, so it's harder (more expensive) to persuade Chinese workers to do low end jobs. Average factory worker wages in Chine are now 5x those in India, because their value-add has rocketed. Chinese workers have better opportunities now.

It's what it is today because that's how it's been adjusted by the capitalists, economists and politicians. I'll avoid the inevitable diversion into neoliberalism.

The fifties and sixties had a sustained period of increasingly shared worker participation, believed in by both sides of the political fence. Rising pay, conditions, pensions, fringe benefits, and reducing inequalities. Regulation to combat the ills of capitalism and markets, to prevent a repeat of 1929, to be suspicious of overly large corporations, to improve the lives both of worker and residents near the factory. It was Republican Nixon who formed the EPA and brought the clean air act.

By placing commerce above nation you restrict the impact of government, the shared deal is defenestrated, and it becomes a global race to the bottom. It makes it super-easy to trade one nation against another, and move profit around to avoid paying the rightful tax. Just because. The corporation is now above its government. Make it wherever's cheapest or has fewest restrictions. The current home of manufacture rises, the globe as a whole does not. Environmentally, climatically and regulationally it got worse.

Not to say I object to other nations rising, I don't at all. It could have been done without throwing baby out with the bath water.

What has happened that is hurting us in the West is that wages have stagnated. Our standard of living is still going up, but that's largely driven by cheaper goods from developing countries (cough China cough) and better technology.

I suspect this is simply due to the fact that there are now hundreds of millions more middle class people in the world that can do our jobs just as well as we can. However I can't bring myself to cry foul. Why shouldn't they get a fair bite at the apple?

Inequality is definitely an issue, but frankly if we're going to vote in billionaires so they can give themselves and their rich mates historically unprecedented tax breaks, while profiting enormously from their position of privilege, maybe we deserve it.

Aye it's hard to talk against your last point. Though did they choose it or were they marketed to vote that way? It's a remarkable success story that so many have voted against their own best interests over the last 30 or 40 years. Though there was a decided lack of alternative views, just degree.

A worse outcome.

Coordination is required to avoid suboptimal solutions to games (as in game theory) that emerge in a modern economy.

If your a computer scientist saying individual freedom resolves all economic questions is like saying greedy algorithms provide the optimal solution to all cs problems.

> Globalisation is fundamentally about individual freedom.

You mean individual freedom for the rich?

> This dynamic has raised hundreds of millions of people out of poverty

And billions of people into modern slavery. And so much more complex misery that it is impossible to comprehend.

> It has increased pollution and accelerated the rate of consumption of resources.

Indeed! We are totally destroying life on this planet and it seems most people don't realize it's the economy. Economists have a 'raise people out of poverty' fallacy for this. All these false truths we learn at school are so persistent, amazing.

I do remember when there was a strong left "anti-globalisation" movement, mostly focused on the "G8" and "G20" meetings. Two things seem to have contributed to ending that: one is heavy police action against the violent demos, and the other is the rise of "rightwing anti-globalisation" which tends to be anti-immigrant and anti-semitic.

Joseph Stiglitz and Naomi Klein are probably the standard starting points for reading about anti-globalisation.

No one seems to understand! Macro or micro economy, for me it's just another pyramid game. For some reason most people are completely unable to see that and can only humdrum what they've learned at school; that 'our' economy is a boon. Well, it might be a boon for the rich, but not for us commoners/slaves. And as a side effect this economy(of waste) is destroying not only billions of lives, but also our entire planet and species.. And we are so smart that we have an opinion whether to choose between Trump or Clinton, while neither of them desire to end this idiocy.

I would not be surprised if we eventually find out humanity has already multiple times wiped everything from this planet in the last billion years. And IMAO we are stupid enough to do it again and again.. It all comes down to: Greed, Vanity, etc..(you know your sins don't you?). The flesh is weak, so weak we'll probably never learn..

No matter how complicated you think economics is, it's more complicated than that. Just the act of trying to understand and theorize about economics changes economics, like an observer effect.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact