« "Social cleansing" and network effects | Main | Some doubts about NGDP targets »

April 26, 2012

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d83451cbef69e2016304cbdd83970d

Listed below are links to weblogs that reference Limits of QE:

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Luis Enrique

what do these arguments mean for those who think all we need is central bank that targets NGDP, and for those who believe the great recession was actually caused by too-tight monetary policy?

talking of "unconventional QE" why isn't government spending financed by printing money more often discussed as an option. I know the standard response is that the BoE just cannot commit to permanently expanding its balance sheet, because its prime directive remains controlling inflation which means it will reverse QE if it need to regardless of any promises.

But if the Bank of England was to say "we've bought all these gilts and we don't necessarily have to sell them again [*] - which effectively means a big chunk of UK govt debt is potentially written off, meaning the govt could afford to cut taxes / raise welfare payments / fund investment. just saying."

and if the government doing so does cause growth and inflation, well then the BoE does end up de-monetizing that debt (reversing QE) but in that scenario (growth + inflation) that wouldn't be a worry.

[*] or fail to roll them over

Luis Enrique

I didn't express that very well. If the BoE encourages the government to borrow and spend, and matches the fiscal stimulus with additional purchases of government debt (more QE) then if the economy stays depressed, the BoE can keep those gilts on its balance sheet, and if the economy is stimulated, the BoE can reverse the QE, which in that scenario won't be a problem because the government's debt capacity will have risen thanks to growth+inflation.

What could go wrong? We could get an outcome with no GDP/employment gains but inflation, leaving the BoE wanting to reverse the QE but the government less able to carry the debt and regretting having agreed to attempt fiscal stimulus. How likely is that outcome?

Andrew

"QE, then, is not enough. So, what else could be done?"

Amazing and frightening that discourse is going on at this level. In my opinion there is not nearly enough humility about the uncertainties of these analyses.

QE equivalent to 20% of GDP raised GDP by about 2% according to one (no doubt very flawed) analysis. And that is proof that it is a net good is it? That's all we need? No analysis of costs? No analysis beyond the short term. GDP is up short term. Relative to what we reckon it might have been otherwise.

Hooray.

Nevermind such concerns, we say. More QE is obviously not "enough". Let's ponder how we can give ever larger and costlier pushes on the economy, with what long term result we have not a clue. But if we can just shove up GDP short term, then short term employment is sure to follow, and that makes long term employment more likely and and and.

What is the product of error on these long chains of assumptions and extrapolations of simple linear relationships far into the future?

No one has any idea.

Andrew

@Luis - "How likely is that outcome?"

No one has any idea. But if the govt is less able to carry the debt then it might be very costly.

Conversely the idea that general fiscal stimulus ever led to sustainable net gain in an economy has not been established.

Keith

"Conversely the idea that general fiscal stimulus ever led to sustainable net gain in an economy has not been established."

The word sustainable is doing a lot of work here!

As a practical matter this assertion is clearly not true. The sustainability of any change in policy depends on the size of the policy change and the size of the problems to be addressed. How long is sustainable in this context? Five years? Ten? Fifty or a hundred? What is abundantly clear is that there is no indication of any significant recovery in output taking place from conventional monetary policy. It is surprising from the historical point of view that Governments are so reluctant to do much of any thing about the fall in output produced by the financial crisis.

May be we could amend the above comment thus-

"Conversely the idea that general private sector credit boom stimulus ever led to sustainable net gain in an economy has not been established." Thus abolishing any connection between bank lending and GDP. But if you agree that we live in a credit money economy then that cannot be true. Output varies with lending/borrowing and sustainability is moot without specifying the details of lending criteria future income streams available to meet the interest etc.

Britmouse

My take: http://uneconomical.wordpress.com/2012/04/26/the-limits-of-uk-monetary-policy/

Also worth noting that if the Treasury decided it really really wanted faster demand growth, they have wide scope to change the definition of "price stability" just by sending a letter to the Bank of England. A 4% inflation target would be the simplest option.

Keith

Britmouse is right about that I think; but no one really seems bothered by demand these days!

Ralph Musgrave

Chris says, “We often think of fiscal policy as being constrained, either by bond markets or by administrative problems.” My answer is s*d the bond markets. Britain is a monetarily sovereign country: it prints its own currency.

For a monetarily sovereign country to borrow its own currency is as pointless as a dairy farmer buying milk in a supermarket. No sorry: it’s even more pointless. Reason is that when a dairy farmer produces DIY milk, that costs the farmer in real terms (labour consumed, etc). In contrast, pounds sterling can be created simply by clicking a computer mouse in the Bank of England. Alistair Darling created £60bn by clicking a mouse at the height of the bank crisis.

As to “administrative problems”, changing VAT or national insurance contributions is as difficult as organising a piss up in a brewery. VAT has been changed twice in the current recession. And if it’s not that easy, then VAT, NI contributions, etc need to be re-structured so that altering them IS EASY.

Luis Enrique

Andrew

I think you are confused. We are not after a"sustainable net gain" in the economy, we are talking about short-run policy to alleviate the recession, reduce the misery and hardship imposed by unemployment. It doesn't matter if 10 years from now the "gains" have evaporated in GDP terms, if we have helped people in the intervening years. Although if you think about the lasting effects of unemployment, then maybe gains are sustained in that sense.

also, it is a nonsense to compare QE (printing money) to GDP (the production of real output) and of course people have thought about the costs of QE - what do you think the costs of QE are? Inflation. That's all the BoE thinks about - trading off the cost of QE against its benefits.

of course there are unknowns in all directions, chains of possible errors as you put it - but that applies equally to not doing QE as it applies to QE. You think the impact of QE is uncertain, the impact of policy passivity is uncertain too.

Andrew

@Luis
I disagree that you have demonstrated confusion.

You are confused. If there is not net gain but rather net loss, to an economy or society then no amount of short term"alleviation"can be justified. People may disagree about the distribution of costs and benefits but any analysis that concludes more harm than good cannot be recommended. If you want to justify qe because of hysteresis effects or long ten social goods then do it. BUT YOU MUST DO IT RIGOROUSLY. Unfortunately what passes for analytic rigor in macro economics would be regarded simplistic in many other spheres. Your following point is an example.

It is not ludicrous to compare the size of qe to GDP, since they are both measured in identical units. If your point were true it would follow that it would be ludicrous to mention its size in pounds.

As to the consideration of the costs of qe, if you think inflation is the only one, you may again refer to the last two sentences of my first paragraph.

It is not nearly as simple a aggregate demand versus inflation. Interest rates versus investment.

Why shouldn't primum non nocere apply to the doctors of the economy?

Luis Enrique

Andrew, I'd tone down your complaints about lack of rigor, glass houses and all that. Comparing the stock market valuation of companies to the GDP of countries is silly, both are measured in pounds. Comparing the volume of money printed to the volume of the UK is silly, both are measured m3. QE is measured in pounds, but the GDP of th country is not a useful comparator.

Its hardly rigorous to claim no amount of shortterm gain can justify a long term loss. How about long term loss epsilon, short term gain a squiliion? In the long run we are all dead. There are millions of unemployed now who need help now.

There is no neutral position, there are only choices. If you deem the case for more QE to lack sufficient rigour, you choose no more QE. But where is the rigorous case for that choice? It's all very well pointing out we live in an uncertain world but that does not amount to a rigorous case of inactivity in the midst of the worst recession since 1930.

What costs of QE do you have in mind, other than inflation?

Andrew

@Luis. I still can't see anything remotely silly about comparing the size in pounds of a qe injection to an economy to the annual turnover (in pounds) of that economy. If I told you that a planned qe was either 0.1% or 100% of GDP that would give you significant information about its likely size of effect.

After all, one of the things we are trying to influence is measured by the GDP!

And you appear to be simply asserting my silliness without any argument.

"Its hardly rigorous to claim no amount of short term gain can justify a long term loss. How about long term loss epsilon, short term gain a squiliion? In the long run we are all dead. There are millions of unemployed now who need help now."

Childish level of thinking. Obviously the long term analysis would still be over a term we actually care about. Whatever the short term gain, if the net effect is negative over that term, then all of that gain and more is surrendered over a timescale we still care about. Obviously, and by definition. Millions of unemployed who need help now is an entirely irrelevant point. The question is not whether there is need, it is WHAT to do about it.

"There is no neutral position, there are only choices. If you deem the case for more QE to lack sufficient rigour, you choose no more QE. But where is the rigorous case for that choice?" And the reductio ad absurdam?:

What is the rigorous case against the infinite number of possible policies we could try? I presume you are familiar with them all, since you apparently recommend not doing any of them, with one exception?

In any case, QE has a number of obvious possible arguments against and if the only cost you can think of is inflation, then I'm afraid you aren't up to snuff.

The mere existence of an emergency doesn't justify "just doing something" on the basis of the facile sort of analyses we get from macro guys in and outside the BOE.

Luis Enrique

Andrew,

OK, fine on the QE thing. I don't think it's particularly enlightening - better to talk in terms of the size of the BoE balance sheet - but if comparisons to GDP float your boat, fine.

As for the rest ... what do you have in mind as a long-term cost over the relevant timescale? Let's say 50 years is relevant and lets fix ideas by saying that actions to address the recession today causes the path of GDP to be x% lower, or the path of unemployment to be x% higher (think of a line on a time series graph and just shift it up or down). So there's a long-term cost. And the short term gain is an X% reduction in unemployment. Clearly there is some x small enough and X large enough for that to be worthwhile.

Or are you arguing that no short-term gain can be justified if the NET effect is negative (i.e. *after* you have traded short-term gains against long-run costs, the net impact is negative) well duh, that's the definition of net cost/benefit.

No I don't have rigorous cases for everything - you're the one complaining about insufficient rigour, I'm the one arguing for policies that are expected to be helpful, on the basis of the best theory and evidence we have at our disposal, even if this theory and evidence doesn't past must with Andrew.

so, are you going to share any of these "obvious" dangers of QE, other than inflation?

The comments to this entry are closed.

Why S&M?

Blog powered by Typepad