Central Banks: Second best policy and operational separation
Following a major crisis, such as the one we’ve just experienced, it is easy to get into a situation where the goal of policy is to “avoid another crisis”. However, this is not a trap that our fine readers fall into – so we don’t need to worry about it here 😉 .
This isn’t to say that we shouldn’t take things from a crisis – far from it. A crisis gives us information about the behaviour of the macroeconomy, about the feed-back loops that may exist that we may not previously pay attention to, and about the process people use for forming beliefs. But we still need to say why these things occur – and hopefully make our given hypothesis testable – before we can decide to do anything.
So well prior to the crisis there was a string of micro-prudential policies introduced. In truth, the retail banks have been grappling with the introduction of many of these policies during the crisis – and the Reserve Bank has stated that part of the reason the OCR is so low is because these policies have, in of themselves, tightened credit conditions.
Now all this is fine, when it comes to cyclical policy the impact of both micro and macro-prudential regulation has been widely discussed. However, what bugs me is the lack of heavy discussion and analysis is the lack of significant discussion around the structural impact of said policy.
I can see a reason for monetary policy, micro-prudential policy, and macro-prudential policy – sure. But my concern is that they are all being looked at within the same frame too much. There are cyclical issues, financial stability issues, and broad structural issues that come out of these types of policies.
What do I mean? Lets focus just on micro-prudential regulation for now. Say that the Bank decides to base some set of reserve requirements for retail banks on the basis of the risk associated with the asset classes they hold – and assume that the Bank SAYS what it believes the associated risks are when setting up the reserve ratio. Well then their weightings on risk have an impact on what bank’s lend out (their asset position) and this in turn changes the availability of funds for different markets, leading to a change in the allocation of investment and general resources.
It becomes more than an issue of “security” – there is a trade-off where normal market signals are being blunted.
In my opinion, the reason we like such policies is because retail banks have a lender of last resort – they will be bailed out in the case something goes wrong, and this leads to a moral hazard problem. The idea of something like “microprudential” regulation is to solve some of the excesses that come from this moral hazard – under the presumption that the “first-best” market alternative isn’t realistically available. In this case, a central bank CAN’T commit to not bailing out banks, and so they regulate them more directly as a means to solve the implied moral hazard issue.
I can agree with this, but there is a cost. It will impact on allocative efficiency (it will change where resources are allocated relative to where they are most valued, for those who don’t like the wonkishness of the term 😉 ). And whatever organisation is responsible for financial stability, the micro or macro-prudential framework, or any sort of structural policy, needs to take responsibility for this.
And that is why I think the function of financial stability and associated regulation should be functionally separate from monetary policy – in the same way that fiscal and monetary policy are separate. Monetary policy and the monetary authorities should solely follow a cyclical frame – other organisations (which they will in turn communicate with, and intrinsically co-operate with) will deal with structural issues.
This is the only way to ensure policy is transparent. It is the only way to ensure the relevant organisation have to take responsibility for their actions and the outcomes associated with them.
I have a few problems with your reasoning:
(1) The lender of last resort role is naturally held by the central bank, not a financial regulator. Is it realistic to expect the central bank to carry that risk, but to leave another agency to determine the quid pro quo for it?
(2) There’s no reason to expect financial regulators, whether independent or not, to be held accountable for the unseen costs of allocative inefficiency. In fact they tend to end up justifying their own existence: regulation stifles competition, which means fat profits for the incumbents, which increases public support for more regulation. And when things go wrong, everyone in power agrees that the cause was a lack of regulation, not a lack of accountabiity.
(3) The case for separating monetary and fiscal policy is not about transparency, it’s about the risk (inevitability?) that the former is used to serve the latter. I can’t think of an example of how that would happen between monetary policy and financial regulation, can you?
Banks have not invested any thing, they cannot have any loses. They do not loan money,
they have only insurance scrip and its of no value. They only make an entry in a ledger to
indicate credit or debt. There is no value in any paper or digital transaction involving USD’s
because of no backing, no real wealth. Under Law, so called bank loans, lack Lawful consideration and therefore are void, see Credit River Township, Scott County, Minnesota
, Jerome Daly v. First National Bank of Montgomery, Judge Martin Mahoney 1968
“I can agree with this, but there is a cost. It will impact on allocative efficiency (it will change where resources are allocated relative to where they are most valued, for those who don’t like the wonkishness of the term ).”
There’s sometimes a cost, some times a benefit. Banks lending out money knowing they will get bailed out will lend in riskier areas than is allocatively efficient, because they don’t bear the full risk of their loans. So microprudential stuff can improve allocative efficiency. You’re right there’s a cost if the regulator pushes too far, but I don’t think it’s right to just say “there is a cost”.
I think you make a good point that the RBNZ will probably be more focused on stabilisation and may over regulate. But wouldn’t the hypothetical new agency focus to much on structural and not consider the effects excess lending has on macro stability? Or are you seperating regulations that are focused on micro problems, microprudential, and macro problems, macroprudential, and then giving them to the proper agency? Is it possible to make that seperation? ie. are there some regulations that are both macro and micro focused?
Another more practical problem is making a whole other agency comes with a lot of costs. We’ve seen the government amalgamate a lot of government departments. For a lot of these you could argue “we should keep them seperate so they have more specific goals”, but the government sees these benefits as not outweighing the costs. Do you disagree? Do you think this new agency would be different from the government departments?
A more practical solution would be to make it explicit that the RBNZ has the OCR and other measures focused on stability and regulations focused (in part or whole) on dealing with moral hazard. Do you think that would be enough?
@Miguel Sanchez
Ello, fair points – lets discuss:
“(1) The lender of last resort role is naturally held by the central bank, not a financial regulator. Is it realistic to expect the central bank to carry that risk, but to leave another agency to determine the quid pro quo for it?”
Yes. I think the discrete choice of bailing someone out should be made by a separate authority that can take funds directly from the central bank – just to make sure that the lender of last resort role is treated separately from the cyclical monetary policy role.
“(2) There’s no reason to expect financial regulators, whether independent or not, to be held accountable for the unseen costs of allocative inefficiency. In fact they tend to end up justifying their own existence: regulation stifles competition, which means fat profits for the incumbents, which increases public support for more regulation. And when things go wrong, everyone in power agrees that the cause was a lack of regulation, not a lack of accountabiity.”
That is an excellent point. But I would note that having a separate authority that has a single, testable, role is more likely to be criticised clearly than one that has multiple roles.
“(3) The case for separating monetary and fiscal policy is not about transparency, it’s about the risk (inevitability?) that the former is used to serve the latter. I can’t think of an example of how that would happen between monetary policy and financial regulation, can you?”
I can understand this but:
1) Transparency is a major point IMO – the whole single instrument-single target logic was for the clear policy link, and I think that institutional structure should follow that. Again, the fact I keep using the word “should” indicates that it is my belief.
2) The central bank is inherently less democratic than government, or fiscal authorities, why do we believe that an undemocratic authority will necessarily perform things in a cleaner, more socially interested fashion unless there actions are transparent and punishable.
3) My example of one “serving” the other would solely be on the basis of attributing blame – if you have multiple goals, you can blame the failure of one on the fact you needed to achieve the other. In the end, the incentive to reach any “difficult” goal is lower.
@Talosaga
“You’re right there’s a cost if the regulator pushes too far, but I don’t think it’s right to just say “there is a cost”. ”
Fair point. When I said their was a cost, I was comparing it to an optimal allocation. Now, of course if the bank faces the incentive to lend too much in general then they aren’t pushing for the optimal allocation – I agree.
However, I was meaning my statement to be in comparison to the “optimal allocation” where moral hazard is not an issue. Why did I do this strange thing? Because I wanted to describe the acceptance of a “allocative efficiency loss” as a second-best outcome relative to one where we just let the moral hazard issue run around.
So the realistic outcomes aren’t as good – but by comparing them to the “first best” outcome, we can say which one is relatively closer, and which one we should do. I was concerned because we are willing to say that we get a benefit by setting up these RR conditions – but it is only when we compare both outcomes to the first best outcome that we get an idea of this allocative efficiency loss. I’m sorry that I was so unclear – I’ve obviously had too much coffee today 😉
“A more practical solution would be to make it explicit that the RBNZ has the OCR and other measures focused on stability and regulations focused (in part or whole) on dealing with moral hazard. Do you think that would be enough?”
When it comes to actually setting policy, I can legitimately understand why a bunch of tools would be used – I would like the case to be made clearly before it happens, but I can understand why.
My main concern is I think there needs to be operational separation between the different tools – as you can tell from the comments so far this isn’t necessarily an agreed upon thing, it is just a belief of mine.
@Matt Nolan
“I think the discrete choice of bailing someone out should be made by a separate authority that can take funds directly from the central bank – just to make sure that the lender of last resort role is treated separately from the cyclical monetary policy role.”
But then does the central bank get a say in whether or not it hands over billions of dollars to another government department? This sounds like adding another layer of bureaucracy to no real benefit.
“I would note that having a separate authority that has a single, testable, role is more likely to be criticised clearly than one that has multiple roles.”
But as I said, financial regulators are rarely criticised in normal times – bank-bashing is politically popular and the costs of inefficient regulation are unseen. Generally the only ones who push back against bad regulation are the ones being regulated – that happens whether or not the regulator is a separate authority. Granted, the banks might be better placed to win a PR war against a combined authority, by running the line that “so-and-so policy will push up mortgage rates”; but that would point towards under- rather than over-regulation.
“The central bank is inherently less democratic than government, or fiscal authorities”
Ditto for financial regulators, that’s why both should be given clear mandates. The RBNZ’s woolly financial stability mandate is a product of how (and when) its legislation was written; the mandate could easily be tightened up without resorting to separation.
“My example of one “serving” the other would solely be on the basis of attributing blame”
Yeah, “blame” is the idea I was grasping at. I can’t imagine a central bank ever being able to blame its inability to contain inflation on the need to clamp down on banks, since the ‘right’ policies in each case would actually be complementary (e.g. the latest research – which I think you linked to a while back – finds that keeping interest rates too low for too long leads to looser lending standards).
@Miguel Sanchez
“But then does the central bank get a say in whether or not it hands over billions of dollars to another government department?”
No.
“But as I said, financial regulators are rarely criticised in normal times – bank-bashing is politically popular and the costs of inefficient regulation are unseen. ”
Hmmm, I’m not sure how much concern I have for the way people view banks in this context – the key point is that the role of any structural regulation separate from cyclical regulation and separate again from monetary policy.
I agree with you that there are definite public policy issues here – but my prior belief is always that we should start with a clear transparent distinction for policy, and that any public policy issues are second order. I can be swayed on this, but as with all beliefs I’d probably have to see something.
“The RBNZ’s woolly financial stability mandate is a product of how (and when) its legislation was written; the mandate could easily be tightened up without resorting to separation.”
I agree that in the absence of separation this is also an improvement – clearly defined mandates are really the key point in all of this. One organisation with a clear mandate would be superior than a separation that is unclear and confused.
“Yeah, “blame” is the idea I was grasping at. I can’t imagine a central bank ever being able to blame its inability to contain inflation on the need to clamp down on banks, since the ‘right’ policies in each case would actually be complementary”
If the impact of policy was solely cyclical then there is a justification for this. However, in the case of a structural impact of policy – if the set of loans being made by banks are being redirected by policy in some way – then there should be a separate body that is responsible for this, so that any misallocations resulting from it can be clearly pinned down.
Errr… I think to take this any further I’d need to know what you have in mind in the way of structural vs cyclical regulations. Cyclical bank regulation hasn’t really been tried, there’s no agreement about how it could be done, and I’m not convinced that it would look any different to cyclical monetary policy anyway. You’ll also need to explain to me why it would require more regulators rather than just more instruments.
“I agree with you that there are definite public policy issues here – but my prior belief is always that we should start with a clear transparent distinction for policy, and that any public policy issues are second order.”
I can understand that as a general rule, but for what you’re specifically proposing I think the public policy issues are unavoidable. The incentives faced by the regulators mean that “cyclical” measures will have a nasty habit of becoming permanent, and that “structural” measures will be dictated by the cycle as regulators try to fight the last battle. The distinction can’t be maintained.
@Miguel Sanchez
“Errr… I think to take this any further I’d need to know what you have in mind in the way of structural vs cyclical regulations.”
Fair call. Sorry I’ve been awful in this post – I had ideas for about 4-5 posts, realised I didn’t have time, tried to write all the things I wanted indirectly in one post, and ended up with a pile of gunk 😉
When I say cyclical vs structural one potential set of policies I would have in mind is the difference between macro-prudential regulation and micro-prudential regulation.
“You’ll also need to explain to me why it would require more regulators rather than just more instruments.”
I don’t envisage more regulators persee – just the “operational separation” of the elements. The same way telecom got operationally separated.
Like I say, its to make the message clear and to make a set group seem “responsible” because there is only one thing placed on them.
“The incentives faced by the regulators mean that “cyclical” measures will have a nasty habit of becoming permanent, and that “structural” measures will be dictated by the cycle as regulators try to fight the last battle”
This is true – but for some reason I think clear separation makes this occurance less likely, while the status quo is likely to lead to policy from central banks becoming incoherent.
“When I say cyclical vs structural one potential set of policies I would have in mind is the difference between macro-prudential regulation and micro-prudential regulation.”
I don’t agree with that distinction. Macro- and micro-prudential regulation are meant to be complementary – in fact, the basis for macro-prudence is the belief that stabilising the parts of the system is not sufficient for the stability of the whole. They have the same goal, namely reducing the tail risk of a costly collapse (with some nod towards efficiency if you like, to keep them from going power-mad). That gives both of them the scope for cyclical or structural measures. Some examples:
Basel capital accord = micro, structural
Core funding ratio = micro, cyclical
Local incorporation rules = macro, structural
Mortgage levy = macro, cyclical
You might quibble with my categorisation, but I’m sure you see the point.
@Miguel Sanchez
Interesting. You are right that the distinction should be between structural and cyclical – not arbitrarily between micro and macro. When I was thinking macro I was thinking along the lines of reserve ratios changing over the cycle – when I was thinking micro I was thinking specifically of the basel capital rules.
In both cases there is a focus on financial stability – and from separate reasons and channels. And I can buy the fact that these rules – both cyclical and structural – are so interlinked with the idea of financial stability that they should all be controlled by one organisation.
However, I still think that this organisation should be separate from the monetary authority – whose goal is essentially price stability, and the smoothing of the economic cycle that is part of that aim.
This way we have two given outcomes, and two distinct authorities that can take responsibility for said action. Their choices impact upon the choice of the other organisation – no doubt – but the separation increases transparency and allows them to help control expectations.
Cool, this is probably a good point to leave off. I think where we may differ is that I don’t feel that people have that much trouble distinguishing between the RBNZ’s two roles. Sure, when it comes to financial regulation the discussion inevitably turns to “what does this mean for my mortgage rate?”, but that’s a perfectly valid question – it doesn’t mean that people are getting it confused with monetary policy.
For the record, I’m ambivalent about separation. But you’re arguing the ‘pro’ and I’m probably the only person here who cares to argue the ‘con’…
@Miguel Sanchez
Yar, definitely agree with you.
Its been a useful conversation – this is the sort of reason I like blogging, it is useful to flesh these things out and to get useful information out of other people 😉