If I am honest, I found Solow’s attack on DSGE a little strange, and fairly inconsistent.  Here is why:

Update: (Before saying why I think this, I should say) This initially came from an Arnold Kling post, and the lingo regarding biology has moved into some of Brad’s negative writing about economists he disagrees with (ht Economist View).  My own view is that these authors are attacking a straw man when they attack DSGE’s (the straw man I’m talking about is the 1980’s RBC model, as I discuss later) – instead of attacking the inappropriate assumptions of some of the practitioners they have decided to attack the method.  This disappoints me.

Yes – all models have weaknesses.  But it is about seeing what to use and where.  And contrary to Brad’s language he is being far from objective in his attack on this type of modeling.

Now, back to the post as it was a few days ago

On unemployment and “the model”

In one paragraph he states that we have frictions such as rigid prices for goods and labour – then in the following paragraph he says that there is no such thing as unemployment in a DSGE model, which is obviously false from the first point.

However, this is merely a small point compared to what I see as his broad critique of DSGE’s, which is that they treat the economy like a couple of decision makers acting optimally, when the economy isn’t like that.

This is a strange criticism.  A DSGE model does have few agents, often with infinitely long lives.  However, these are simplifying assumptions – and many of the results can be generalised to heterogeneous agents who have finite lives (but over-lapping generations of people who value their descendants).  There is definitely value in researching the impact of heterogeneity and the such – and that is exactly what macroresearchers appear to be doing – at least judging by the mailing lists I’m on for macro papers.

Furthermore, what is Solow’s alternative.  Is he suggesting that me model each individual in society?  I don’t see how that could provide tractable results, or even if it is remotely possible.

On the use of the model

What ultimately kicks me is the conclusion, which is absolutely at odds with what people use DSGE models for!

The point I am making is that the DSGE model has nothing useful to say about anti-recession policy because it has built into its essentially implausible assumptions the “conclusion” that there is nothing for macroeconomic policy to do

This is abjectly false.  The scope for policy intervention relies on the assumption of “rigidities” which are implicitly proxies for the very market failures he is begging for from economic models.  There are a swath of papers coming out of universities every year using DSGE modeling to justify all sorts of policy interventions, and justifying all sorts of responses to economic events.

The problem isn’t that the model always tells us to do nothing – the issue is trying to fish through the differing assumptions that are being made about the aggregate economy, and trying to find which assumptions we think are most appropriate for discussing the real economic situation.

You seem confused

I am a touch confused – as Solow was effectively the godfather of this form of modeling.  His long-run growth theories were one of the major building blocks of what has become DSGE models.  As a result, the man does know what he is talking about.

But his critique of DSGE models that

1. They don’t allow for unemployment and
2. They immediately dictate that there is no role for stabilisation policy,

Doesn’t seem to follow from the current set of DSGE modeling, but instead from critiques of old school Real Business Cycle theories from the 1980s.

Is there a lot of work still to be done on DSGE models to ensure that we have a clearer description of the economic situation, yes.  Do practitioners need to be mindful of the limitations of DSGE models when forming policy, yes.  But in no way does DSGE modeling act like all unemployment is voluntary and that there is no role for government.

Now, I am assuming that the critique is about the use of DSGE models to form policy and describe the economy – if the critique is actually about the predictive power of DSGE models I think the case is even weaker, as predictive weakness is an issue that holds for the entire discipline – and it is not a game killer.

The reason why I find this way of looking at tax changes strange is that it ignores how prices change in response to the structure of the tax system.  I fear that, to many people, this seems like a benign (possibly even esoteric) issue – when actually it is one of the most essential issues to keep in mind when thinking about the design of a tax system.

Partial equilibrium

The first step required to analyse this is our good friend the “economic scissors” – or supply and demand.  The introduction of a tax creates a “wedge” between our supply and demand curves, which in turn implies that the quantity produced in the industry is lower, and the prices faced by the supplier and the purchaser are different.

Now in the case of the labour market, the purchaser is a firm and the suppliers are workers.  Only by looking at how responsive supply and demand are to changes in the “price” (wages)  can we infer what will happen when we change the tax rate – both to the welfare of the individuals in the market and the output produced in the market.

Update:  Eric Crampton discusses experimental evidence that shows where people struggle with this concept.

General equilibrium

However, our scissors only take us so far.  The labour market is not independent of other markets out in society.  A higher tax in the labour market implies that input costs are higher in the goods market – a factor that influences prices and output.  Furthermore, lower production in the labour market implies just that – lower production/output.

Given the general nature of a tax on all labour income, and given the fact that labour is an input, it is not really clear exactly how a tax on one group will impact on everyone is society.  It is possible to say that, although a tax on the wealthy which redistributes will make the poor relatively better off – it may also make them worse off in an absolute sense [note, this is not inconsistent with the second welfare theorem as a labour tax has a supply side impact that a "redistribution of endowments" does not].  Now I don’t believe that we are near a stage where this is the case – but it does illustrate that the idea of redistribution is inherently fraught with complications.

Time path

Furthermore, the welfare costs of the transition between different tax systems are inherently difficult to understand.  Saying that so-and-so gets a $Y a week tax cut may be true in that very moment of time – but within a year the path of prices in the goods market, the negotiation of wages in the labour market, and the path of government spending will all have likely changed relative to their pre-tax cut level.

As a result, how this transition functions has costs and benefits in of itself – which is also an important issue to understand.

Conclusion

As we have described above, the issue of “redistribution” through “labour tax” is unclear.  This is one of the primary reasons why economists like the idea of a “flat tax” – since the redistributive impact of progressive taxation in itself is unclear.

If we combine a flat tax system with a targeted benefit system (which is based on both extensive research and an exploration of societies preference for redistribution) we have a clear and transparent equity-efficiency trade-off.  This seems like an appropriate way of forming, and then fulfilling, our social contract.

However, the same people complaining about the $Y a week so-and-so gets would criticise such a shift to a flat tax system on the same grounds.  Hopefully, by pointing out that such a static view of tax changes doesn’t make sense we can help put the desires for a more transparent tax system in context.

We need to design the tax system so that it allows the government to take the money it requires, while doing the least amount of damage to the economy and so too our potential prosperity

However, there is a proviso that needs to be taken into account when we say this.  Any redistribution that we as a society deem is appropriate given our value judgments needs to occur through “the money the government requires”.

The tax system that is solely based on efficiency will not be a tool for redistribution.  Depending on our value judgments, we will want a certain level of redistribution, and this has to occur through the level of government spending.  The higher redistribution is, the higher government spending is, and as a result the higher the tax rate will have to be.

Yet, according to this post, the recommendations of the report switch from the design of optimal tax to the equity-efficiency trade-off associated with redistribution:

A 2001 OECD study found that about one half of a percentage point increase in government consumption (the expenditure to GDP ratio) could cause a 0.6 to 0.7% direct reduction in per capita output.

Yes, there is a trade-off.  However, the level of government spending and redistribution should be premised on this trade-off.  By saying something like “If we can limit spending so that over time it is under 30% of GDP” we are making a value judgment regarding the amount of redistribution that is in societies interest – we aren’t discussing the role of optimal taxation.

My main point here is, there are two separate issues:  Firstly, the design of an optimal tax system GIVEN the level of redistribution.  Secondly, the socially preferred level of redistribution.  The first question is easy, even an economist can answer it.  The second question is incredibly difficult.

Source.

This is one of the reasons why taxing on the basis of a “family” unit instead of an “individual” unit doesn’t make sense to me.  By setting up an arbitrary idea of what a family is you ensure that people arrange their affairs to take advantage of that, and you ensure that people that are either unwilling or unable to enter these arrangements struggle.

People pool resources and work together by forming a family unit if they want to.  Lets just stick to taxing people as individuals, and let individuals in society make the decision on what type of family unit to set up given this equal treatment – after all it is the individuals that make choices.

(Economic) predictions are, of course, the bread and butter of economic institutions. But can we believe them?

In recent years, some economists have begun to express doubts over predictions made from huge volumes of data, but they are in the minority. Most embrace the idea that more measurements mean better predictive abilities.

Hold up.

For one, as we have mentioned prediction is not the central element of what economists do – and even when they do predict the goal of such prediction is to give some view regarding risks and movements, not direct figures (it is more ordinal than cardinal in some sense).

Secondly, ever since the Lucas critique economists have been very nervous about predictions from large amounts of data without theory – I would say that the majority of economists doubt the usefulness of econometric models relying solely on huge amounts of data.

Economists would like data with less measurement error, that is closer to representing the true economic variables we discuss in theory – we aren’t looking for an infinite number of measures we can stick together to find a result.  An economist that doesn’t use theory to inform their discussions of the economic outlook, but uses lots of data, isn’t an economist – that is all.

However, in the same way I don’t believe the difference in these games is just the product of “framing”, I am not sure if the call to arms against using interest rates as a focal point is necessarily that compelling.

On quantity and price competition

Back in the 19th century some economists called Cournot and Bertrand came up with separate models of firm oligopoly behaviour.  In the Cournot model firms picked quantities, and they kept some market power – albeit less than in the monopoly case.  In the Bertrand model firms picked prices – and we got this crazy result that merely having two firms in a market provided perfect competition.

As a result, economists became concerned.  We had two models, one which seemed to fit data better (Cournot) and one which had assumptions we felt were more realistic (Bertrand).  A multitude of ex-post imperfections could be introduced to a Bertrand game to create “supernormal profits”, such as heterogeneous goods, transaction costs, and imperfect information – but there was still the problem that, if we had only two firms competing and they decided to “play” in prices instead of quantities we had a different outcome.  Given that the demand curve is the thing along which both price and quantity were picked, and given that the firms in both cases were seen as equivalent, this didn’t seem consistent.

Eventually economists realised that there was something else at play here.  The Cournot and Bertrand firms were not equivalent at all.

In the Cournot game the question is:  if I LIMIT myself to producing a quantity how will other firms react – and given that reaction what level of quantity would I want to limit myself to.

In the Bertrand game the question is:  I have an unlimited ability to produce.  As a result, if I set a price how will other firms react – and given that reaction what price would I charge for the produce I will be able to make.

So in the Cournot game you build things first, set the price later.  In the Bertrand game you set your price and immediately satisfy this demand.  As a result, economists realised that the Cournot game was simply a Bertrand game with capacity constraints.

What in the hell does this have to do with the point on monetary policy!

I was getting there.  Fundamentally, in the same way that Cournot and Bertrand games were discovered to only give different results because they were fundamentally different, I believe that any difference between an interest rate target and other targets only differs if “effective policy” is different – I don’t believe in a framing issue persee (although framing explanations can be funky).

Now I don’t disagree with the idea that just talking about the nominal interest rate would be silly.  But central banks don’t just talk about a nominal interest rate – they also discuss an inflation target, which anchors inflation expectations.  In essence the current “focal point” for policy is the real interest rate.

Furthermore, since they control a real interest rate, and have anchored inflation expectations, they can print money which in turn increases demand for goods and services.  As Nick states:

And most of the power of a central bank comes from its ability to influence people’s expectations of the future. Like governments, police, armies, and referees, most of central banks’ power comes from belief in their power

The fact that inflation expectations are anchored implies that people believe they know the future price level, given that a significant portion of any nominal increase in income will be confused for real income – leading to extra spending activity.  That is the very power of an inflation target – an inflation target that is easy to communicate and explain to the public by discussing interest rates.

Targeting arbitrary variables that are positively related to an economic recovery, but not appropriately related to “monetary policy” seems both sort of aimless and potentially dangerous.  A central bank can keep discussing interest rates and its inflation target, and even at a “zero bound” it could stimulate activity by printing, printing, and printing.

Finally, I think it is important to note that this subject only really matters in the rare occasion that it really matters .  Outside of zero interest rates and a steep, depression like, drop in demand the interest rate (or some indicator of “monetary/financial” conditions) is an amazingly effective tool for communication.

However, my belief is that this effectiveness continues even in the extreme conditions.

• From the Economist,
• From Reform, in the UK.

The Standard has suggested that similar comparison in NZ could be a little out of whack, and in the most part I agree with them.  After all, Labour was elected to a third term on the promise of larger government, National was re-elected to keep it at that level, as a result I think society is suggesting that they want government to continue spending a quarter of our income.

However, I do disagree with them with regards to the idea that government spending didn’t markedly rise as a proportion of GDP in Labour’s third term – to me the GDP Statistics seem to suggest this was the big mover (with the recent increase solely the result of a recession, and “automatic stabilisers”):

Priorities. Obama’s fiscal stimulus did little bar increase the deficit massively, and turn the country into deficit hawks. Unemployment went well beyond his worst forecasts

Now I found this statement unusal in that David’s writing is usually very balanced, and yet I do not find this statement balanced at all.  Why?

1. We have no idea if Obama’s fiscal stimulus did anything until the data is all finalised.  In a couple of years researchers will be able to look over the data and discuss the design, implementation, and need of the scheme and reach an educated conclusion.  At the moment people can only present an opinion on the basis of ideology.
2. Personally (going onto my ideology ), I think the fact that unemployment rose even further than expected was the result of the shock being larger than expected (and areas of the US economy being more fragile).  During the crisis the US government was able to borrow cheaply and use this borrowing to undertake investment when the cost of building this investment was cheap (thanks to the spare capacity in the economy) This sounds like a good thing to me …
3. Unemployment as high as  10% indicates to me that there was a hole in demand – I do not believe that “structural” economic issues could be sufficient enough to warrant 1/10 people who want a job not being able to get a job.  With the Fed unwilling to soften its monetary stance further the government is in a position to be “consumer of last resort”.  Although I don’t really like the idea of this, in the face of sticky prices and a massive shock to the economy I have to concede that such a role exists in extreme circumstances.

As a result, if I had to guess I would say that the immediate crisis would have been worse if the stimulus hadn’t happened. This appears to be a moderate position among economists, between the “stimulus did nothing” and the “we needed more stimulus” extremes.

Now, we may find that the long run impact of this borrowing will be bad, and we may look back on the evidence and find that the scheme is flawed.  However, the point that “Obama’s fiscal stimulus did little bar increase the deficit massively” is an extreme view (that could potentially turn out to be true) – not an objective fact.

However, there are some issues I would like to discuss – lets start with the idea that we “urgently need to cut the corporate tax rate” if Australia does.

Currently there is talk that, if Aussie cuts the corporate tax rate to 30% we need to do the same immediately.  We are told this as if it is a self-evident truth, and told that if we don’t all investment will head to Australia.  This is a touch over the top.

In Australia and in New Zealand there are a whole bunch of “potential investments” that offer varying rates of return.  Now, New Zealand is a small open economy, so we have access to all the credit we want from overseas – as long as the rate of return on that credit (adjusted for risk) is equal to the global rate of return.  So the “supply curve is perfectly elastic”.

A higher corporate tax rate in New Zealand reduces the profitability of investment, which in turn implies that the rate of return for each project is lower.  As a result, a higher corporate tax rate (in a sense) shifts the demand for capital to the left.

As a result, when setting the tax what matters is the “elasticity of demand” for capital investment in New Zealand – not the tax level in Australia.

Will a higher corporate tax rate reduce investment, yes.  It does so by preventing “marginal” projects from occurring – those that offer the lowest benefit.

But it must matter somehow!

The corporate tax rate overseas matters insofar as it changes the “global rate of return” on investment – so if all other countries cut their corporate tax rate, the required rate of return would be higher.  As a result, foreign tax arrangements do impact on what rate of return we have to achieve to get foreign investment.

However, the welfare impact of the tax, and whether we should cut the corporate tax rate depends on the elasticity of demand for investment – it is not a fact that we should attempt match Australia’s tax rate.

To horrendously oversimplify the positions in order to make this post easier to tie together, Sumner seems to state that the Fed needs to print money with a nominal GDP target in mind, Woolsey suggests that the Fed should change inflationary pressure given a set real GDP target, and Kling states that we only have real shocks and so the idea of a “nominal shock” is not of use.

These three views as regards nominal GDP seem to be:

1. Sumner:  The Fed can change nominal GDP and keeping growth in this stable is good
2. Woolsey:  Nominal GDP is pre-determined (within some bounds) and so monetary policy can be adjusted to keep us on a “real growth” path (which is related to potential)
3. Kling:  Readjustment leads to shocks in real output.  Nominal factors are not important

The problem is that none of these theories are absolute – they are all right in part.  Why is this a “problem”?  Well the debate seems to view them as mutually exclusive when in reality bits and pieces of them are happening all the time.

I would say that a stable path of nominal GDP matters because people form expectations on the basis of nominal values.  This is similar to arguing for a stable path of inflation, it just takes it a step further when output exogenously declines.  Given people form contracts on the basis of future expectations regarding prices and the level of nominal activity it makes sense for the monetary authority (who do control nominal variables) to make the adjustment path as stable as possible.

On the other side, Woolsey effectively points out that this idea doesn’t give a clear transition path from nominal to real GDP.  If the Fed targeted nominal GDP instead of inflation, then the inflation target will be more variable.  A more variable inflation target creates uncertainty which could lower real output.

And finally, the idea of nominal targeting cannot explain why output may adjust.  A supply side (or recalculation style) shock such of that provided by Kling could provide this.  Of course, there are a number of other shocks that may do the same (eg undue tightening by the Fed, like bringing in interest on reserves when the optimal interest rate was already hitting at zero.  Or an exogenous, self-enforcing, change in expectations among agents).

Conclusion

In truth we will almost NEVER be able to clearly define whether we are facing a supply shock, a relative price shock, a macro-economic co-ordination failure, or a straight nominal shock.  Why?  Because we will always be facing a little bit of all these things.

What we should be doing is trying to understand the different transition paths these shocks work through.  If we understand how a shock functions, we can be sure to make better policy to limit the cost to society of said shock.

If anything, this indicates that using a DSGE style model (possibly with some room for non-linearities) IS the way to move forward – contrary to all the abuse mathmatical modeling has put up with in recent months.