UK EU Referendum Aftermath

[sadmanbarty]

1) You're making two related points, aren't you, Sadmanbarty? One is that the government is better off taking the advice of economists. The other is that economics is a science, more or less - or at least something close to a science (and so should be trusted by the government, etc, etc).

The first is debatable. 2) To debate it, we'd need to know what the alternatives are. 3) We'd also need to to disaggregate "we" (since it's likely that some people would be better off and some worse off under any economic regime change). Then there's the question of non-material benefits.

4) The second is pretty obviously not true, I'd say.

1) I'd say that's accurate. I'd add that I by no means think that it's a precise science, I appreciate that there are problems in methodology (such as the inability to have controlled experiments) and there are many other fields (particularly in the physical sciences) that comply much more with ideal scientific methodology. Nonetheless economics does use scientific methodology; empirical data is used to test hypotheses, repeated outcomes are sought after, studies are peer reviewed, etc.

2) We have examples of governments pursuing policies that comply and don't comply with the economic consensus. We even have examples of new governments coming in, or governments back peddling, which shows how different approaches work in the same economic context.

3) I was talking about aggregate benefits (though something like Brown's fiscal stimulus benefited pretty much the whole economy). Things like the '81 budget and post-2010 austerity disproportionately hurt the poor.

4) The reason you gave for this was the profession's inaccurate forecasting. There are other scientific disciplines such as medicine or climate science that have these troubles, I presume you still count those as sciences.

 

[vimothy]

I appreciate that there are problems in methodology... and there are many other fields... that comply much more with ideal scientific methodology..

But medicine and climate are both sciences, and so therefore economics is as well? I don't know anything about either field, but I can stil see that that reasoning is faulty.

Sciences proceeds by facts, established by the scientific method. Can you give me some examples of scientific facts, established by economics?

 

[sadmanbarty]

1) But medicine and climate are both sciences, and so therefore economics is as well? I don't know anything about either field, but I can stil see that that reasoning is faulty.

2) Sciences proceeds by facts, established by the scientific method. Can you give me some examples of scientific facts, established by economics?

1) The argument you've put forward so far against economics being a science is that forecasts are inaccurate. My point is that there are branches of science, such as medicine and climate science, that also forecast inaccurately, yet you only discount economics and not those other branches. In other words, if we were to say that a discipline is only a science if it can forecast accurately, we would have to say that climate science and medicine aren't sciences. If not then your arguments so far don't disprove economics as a science.

2) There are plenty of scientific theories that aren't facts. Is black hole theory not science because one has never been observed? Was the 50 odd years of theorising before the Higgs Boson was 'discovered' (assuming it has been) not science?

Nonetheless there are economic facts, for example business cycles (GDP fluctuating between expansion and contraction).

 

[Mr. Tea]

My 2c worth:

All science is inherently empirical, whether that empiricism takes the form of laboratory experiments, environmental observations, clinical trials or whatever. Some branches of science are rather divorced from direct empiricism, but even the most abstruse realms of theoretical physics are still only physics (as opposed to pure mathematics) inasmuch as they could be tested in principle. For this reason, there's an increasing feeling among physicists who don't study string theory/M-theory that this branch of physics shouldn't be regarded as physics/science at all, because it's hard to imagine how it could ever be tested. The same could be said of the Higgs mechanism when it was proposed 50+ years ago, because experimental particle physics was in its infancy then and the machines available were nothing like powerful to create the particles or sensitive enough to detect them. But now they are, and they have been.

Black holes have never been directly observed, true, but secondary phenomena have been observed many times that can't realistically be explained by any other type of object, so their existence is generally accepted.

How this relates to economics, I'm not sure. I would tend more towards the position that you're doing proper science if you can build a theory, or a model of a theory, that's falsifiable; that can be empirically proven not to be correct. Is this possible in economics, vim (in your opinion)?

 

[vimothy]

The argument you've put forward so far against economics being a science is that forecasts are inaccurate.

I thought that you were arguing that economics is a science and I was disagreeing. If economists could predict the behaviour of the economy, this might suggest that economics rests on some sort of body of scientific knowledge (although it's not the same as proving that economics is a science, it's at least suggestive). But economists are hopeless at predicting the behaviour of the economy, and macroeconomists are most hopeless of all.

That doesn't mean that economics isn't a science (logically speaking). But what is the *actual argument* that is meant to establish that is?

 

[vimothy]

Nonetheless there are economic facts, for example business cycles (GDP fluctuating between expansion and contraction).

Cyclical GDP fluctuations at business cycle frequencies is an empirical regularity observed in some time series. It's not a fact established by science. What's the process that generates these regularities? Economists don't know.

 

[vimothy]

Is this possible in economics, vim (in your opinion)?

In general, no. Perhaps there are some subfields of micro where you can do controlled, repeatable experiments, but that's not the case for the most part.

 

[sadmanbarty]

How this relates to economics, I'm not sure.

I was hoping you'd jump in.

My point was that Vim's critiques of economics as a science can also be applied to disciplines that I presume he would consider science, thus rendering those arguments wrong.

 

[sadmanbarty]

1) If economists could predict the behaviour of the economy, this might suggest that economics rests on some sort of body of scientific knowledge (although it's not the same as proving that economics is a science, it's at least suggestive). But economists are hopeless at predicting the behaviour of the economy, and macroeconomists are most hopeless of all.

2) That doesn't mean that economics isn't a science (logically speaking). But what is the *actual argument* that is meant to establish that is?

1) If climate scientists could predict the behaviour of the weather, this might suggest that climate science rests on some sort of body of science. But climate scientists are hopeless at predicting the weather.

If doctors could predict the when people will die, this might suggest that medicine rests on some sort of body of science. But doctors are hopeless at predicting when people will die.

Beside Friedman and Phelps predicted stagflation. Economists predicted the effects of austerity, fiscal stimulus, qe, etc. There are countless examples.

I think the analogy with medicine is useful. A doctor won't be able to predict your disease and likewise an economist won't predict a deleveraging. Yet that doesn't mean they don't know how to respond to those situation.

2) Sorry, I don't understand this. Could you explain please?

 

[Mr. Tea]

1) If climate scientists could predict the behaviour of the weather, this might suggest that climate science rests on some sort of body of science. But climate scientists are hopeless at predicting the weather.

Climate is not weather. We have meteorologists for predicting the weather in the very short term, and they do a reasonable job of it, usually. The accuracy rapidly drops off with forecasts more than a few days, and after a couple of weeks are scarcely more useful than looking up seasonal averages. This is an unavoidable consequence of the effectively infinite number of variables involved and the nonlinear nature of the equations involved in fluid dynamics.

Climatologists have been warning for decades that increasing CO2 levels in the atmosphere will mean a continued increase in average global temperatures, which is exactly what is happening. If anything, many of the predictions from 10-20 years ago have proven rather conservative.

 

[vimothy]

I mean, what's your argument for why economics should be regarded as a science?

What you're doing here is making analogies: economists can't forecast, neither can climate scientists; climate science is a science, so therefore so is economics. That's not a very good argument.

 

[vimothy]

Also, as I said, I don't know anything about climate science or medicine, so I'm in no position to evaluate whether that's true, even if it mattered.

 

[sadmanbarty]

Climate is...

Just to be clear this isn't my real opinion, I'm just pointing out the flaw in Vim's logic. Earlier in the thread I posted how climate scientists had got some predictions wrong, which is what I'm referring to.

 

[sadmanbarty]

I mean, what's your argument for why economics should be regarded as a science?

I would say what constitutes science is by and large determined by methodology. The fact that economists can then apply this body of knowledge to the real world and replicate results also falls in economics' favour.

 

[sadmanbarty]

Also, as I said, I don't know anything about climate science or medicine, so I'm in no position to evaluate whether that's true, even if it mattered.

If you're possibly willing to say that medicine and climate science aren't sciences as well, then I have no qualms with you dismissing economics too.

 

[sadmanbarty]

What you're doing here is making analogies: economists can't forecast, neither can climate scientists; climate science is a science, so therefore so is economics. That's not a very good argument.

I suppose we keep putting the burden of proof on the other person.

I say that economics is a science because it uses scientific methodology to gain results that are repeated when applied through policy.

What is your argument why it isn't a science?

 

[Mr. Tea]

SMB, can I ask what makes you say climate scientists can't predict anything?

 

[sadmanbarty]

SMB, can I ask what makes you say climate scientists can't predict anything?

Again, just to clarify, I don't actually think that. I was taking Vim's critique of economics and applying it to another science. I was referring to these, which I posted earlier:

http://www.telegraph.co.uk/news/ear...dmit-global-warming-forecasts-were-wrong.html

http://www.washingtontimes.com/news/2016/apr/7/climate-change-models-wrong-predicting-rain-drough/

Just to be clear I think that climate scientists, medical professionals and economists all know the most about there respective fields and should be trusted when it comes to implementing government policy. I'm in no way a climate change denier.

 

[vimothy]

You can't prove a negative, but it isn't a science in the same way that sociology or history aren't sciences. It doesn't use scientific methodology (obviously impossible in principle in macro and related fields) and it doesn't generate results that are repeatable or that everyone accepts as definitive.

 

[sadmanbarty]

You can't prove a negative, but it isn't a science in the same way that sociology or history aren't sciences. It doesn't use scientific methodology (obviously impossible in principle in macro and related fields) and it doesn't generate results that are repeatable or that everyone accepts as definitive.

Maybe you could give a definition of science, but other than that I think that we've probably come to the end of this discussion (though I'm more than happy to go on if something else comes up).

To compare history to economics in terms of scientific rigour is a bit ridiculous.

Your qualms with the methodology (as I understand them) are also true for lots of disciplines considered science such as huge swathes of astrophysics. I don't know if you say those things aren't science either.

Similarly there are a lot of things in science that scientists argue over or don't find definitive, economics isn't unique to that. There are things that there is a (sometimes overwhelming) consensus about.

I've given examples of applied economics that produce the expected results.