Winner of the New Statesman SPERI Prize in Political Economy 2016

Sunday 17 November 2013

Should fears of financial instability raise interest rates?

Inflation is significantly below 2% almost everywhere. In the US, Japan and the UK (even though in the UK inflation is still just above 2%) central banks are doing a great deal to get inflation back to 2%. Maybe not enough, but their goal is clear. The ECB is belatedly following the same path (although it remains somewhat behind), but this has caused a very public split in its ranks. One reason given by those who have opposed the ECB’s latest rate cut is a risk to financial stability, and house price increases in certain Eurozone cities. [1] In the US some have raised concerns that continuing QE might generate financial instability. In the UK one of the three ‘knockouts’ to forward guidance, that could allow interest rates to rise even if unemployment remained above 7%, concerns financial stability.

And in one country, Sweden, the independent central bank has kept interest rates above the ZLB, even though prices have been literally falling. While the central bank cut short rates to 0.25 in 2009, during 2010 they were increased to 1%, and during 2011 to 2%. They have since been cut to 1%, but the central bank does not want to cut any further despite prices being flat or falling throughout 2013. Yet the central bank has a clear target for inflation of 2%.

The reason the Swedish central bank - the Riksbank - is overriding its inflation target were clearly set out in a speech given by Kerstin af Jochnick, First Deputy Governor of the Sveriges Riksbank, in January 2013. The Riksbank is concerned that low interest rates will exacerbate a housing bubble. The discussion is interesting for at least two reasons. The first is that the Riksbank is not primarily concerned about the impact of any bursting bubble on the financial sector itself. It is not worried about a second financial crisis. Instead it is worried about the impact a bursting bubble might have on households (creating another balance sheet recession) and overseas confidence.

The second is that the Governor is explicit that higher interest rates are a second best solution to this problem. The first best solution is macroprudential regulation. But, to quote the speech: “One reason why monetary policy in Sweden has needed to give consideration to financial imbalances is because there has been no framework for macroprudential policy.” The speech discusses the progress that the central bank has made in developing a framework for macroprudential regulation.

In passing I would want to add that interest rate policy is probably the third best solution to housing market concerns. The potential for using particular fiscal policy instruments is often overlooked. For example, just as the UK government has tried (somewhat opportunistically) to stimulate the housing market through fiscal means (Help to Buy), these means can also be used to dampen that market. Indeed Goodhart and Baker have argued that Help to Buy can be seen as a macroprudential instrument.

My more substantive point is that a monetary policy of the Swedish kind risks undermining the legitimacy of independent central banks. As regular readers will know, I believe strongly that there are areas of macroeconomics where delegation can be highly beneficial. You only need to look at the influence that misguided, and sometimes crazy, macroeconomic ideas can sometimes have among politicians to see why. On the other hand, delegation potentially undermines the democratic process. There is nothing which says that central bankers, or experts who sit on monetary policy committees, have any particular right to take decisions which can have a substantial impact on people’s lives.

That is one reason why Alesina and Tabellini [2], among others, stress that successful delegation happens when there is a broad consensus on what constitutes sound policy. I think one reason that delegation of monetary policy to central banks has been largely uncontested so far is that this consensus existed for monetary policy. Essentially the task of central banks was to keep inflation low. Of course there is plenty of scope to discuss the details of how this is done, which macroeconomists spend a great of time doing. But the primary task, and the proximate means by which it should be achieved, were clear and commanded near universal support.

For some time the only potentially competing goal was keeping unemployment low: hence the dual mandate in the US. However there was near universal agreement amongst economists that the only sustainable level of unemployment or output that monetary policy should try to achieve was precisely the level that kept inflation stable. If that level of unemployment was too high, then means other than monetary policy should be used to address that problem. Again there are disputes at the margins, particularly when supply or cost-push shocks hit, but little dispute about the basic idea.

The moment central banks start allowing inflation to be persistently below target (with the loss in output that this implies) because of concerns about housing bubbles, this consensus will evaporate. Again Sweden provides a clear example. Lars Svensson, a highly respected academic and a former Deputy Governor of the central bank, has strongly disputed that this policy will achieve the goals it is designed to achieve, and instead suggests that the central bank is violating its mandate. (A good summary is here.) On the more general issue of how much monetary policy should take account of financial risk or housing bubbles there is a wide spectrum of views among economists.   

It is also not difficult to see how reasoned debate could easily escalate into attacks on central bank independence itself. If the central bank begins to be perceived as protecting the interests of the financial sector rather than the public at large, then demands for the government to take back the control of interest rate setting could become difficult to resist (even if the government wanted to resist). I would have no difficulty writing the slogan myself. First the banks created the recession, and now (through the central bank) they want to take away the recovery.

One response to this argument might be that the public would not forgive a central bank that allowed a second financial crisis to develop. I would agree that in the absence of any other remedy interest rate policy should be influenced by the possibility of a financial crisis, as Michael Woodford has demonstrated formally (see this post). So how do you exercise this option of last resort, but still ensure the legitimacy of independent central banks by focusing on the control of inflation? I quite like the arrangement in the UK, where there is a separate Financial Policy Committee (FPC) that works with but is independent of the Monetary Policy Committee (MPC). The FPC, not the MPC, is in charge of macroprudential policy. The knockout to forward guidance that I mentioned above involving financial instability is called by the FPC, and the MPC can then decide whether to act on that call.

Such an arrangement works best if both committees are populated by experts from outside the central bank, and in the case of the FPC those experts are not just current or past bankers. It has the advantage that before the MPC can even begin to consider allowing fears of financial instability to influence its interest rate judgement, the FPC has to in effect say we have exhausted all the other means at our disposal. It would be also good if the FPC was explicit about any micro fiscal issues that might also be involved, just as Bernanke has been explicit about the problems macro fiscal policy has caused him in the last year or so. This institutional arrangement makes it clear that it is for others, and not those who set interest rates, to protect mortgage borrowers from their own potential folly.

[1] See this article by Hans-Werner Sinn for example. To be fair housing is only briefly mentioned there, and the main point seems to involve something else, although what exactly is less clear to me.

[2] Alesina, A. and Tabellini, G. (2007) “Bureaucrats or politicians? Part 1: A single policy task.” American Economic Review 97: 169–179.



  1. So the Riksbank allows excess unemployment so as to control a potential housing bubble: just brilliant. Unemployed Swedes will be deeply grateful to the governor of the Riksbank.

    This whole problem stems from using interest rates to regulate demand: you daren’t cut rates in case mortgagers get over extended, and then get into trouble when interest rates rise.

    The solution is to use fiscal measures to regulate demand, and as to the problem that fiscal stimulus, i.e. “borrow and spend” might raise interest rates, that can be dealt with by having the central bank stop any such rate rises. Hey presto: demand is controlled, while interest rates are more stable.

    That’s the solution advocated by Positive Money. See:

    1. I think your logic is correct. Just as countercyclical policy should be used when interest rates hit the conventional lower bound, it should also be used when monetary policy is constrained in some other way. I would be interested to know if anyone has been making this argument in Sweden.

  2. When Atlee nationalised the BoE, I believe the Labour government then did nothing else with it. But when the Thatcher Cabinet did their monetarism 1980-3, the nationalised BoE gave them full support.

    Is it too odd to think that independence of central banks really does make that much difference, and the are wider problems at play?

  3. You seem to be strongly in favour of central bank independence, although you say "On the other hand, delegation potentially undermines the democratic process."

    Axel Leijonhufvud argues "The independence doctrine, however, is predicated on the distributional neutrality of central bank policies. Once it is realized that monetary policy can have all sorts of distributional effects, the independence doctrine becomes impossible to defend in a democratic society. "

    Can or should Central Bank independence be maintained at a time when consensus on monetary policy is far less strong than a few years ago, and when unorthodox policies like QE with significant distributional effects are being implemented or considered?

    1. I think Leijonhufvud is wrong. Indeed, I think that it is partly because optimal monetary policy has distributional consequences that delegation is useful, otherwise partisan politicians start deviating from the optimum to please particular groups.

    2. Simon, I don’t get the argument that “because optimal monetary policy has distributional consequences that delegation is useful..”. You seem to suggest that distributional consequences are desirable. Surely distributional consequences are by definition POLITICAL in nature. A classic example is the tendency of QE to benefit the asset rich.

      That problem is solved by Richard Werner, Positive Money & Co by having the BoE MPC or some similar committee of economists decide the SIZE OF any stimulus package, while the actual types of spending (and/ or tax cuts) to which that stimulus is directed is left entirely to politicians.

    3. You cannot do that for interest rates. In a recession you need to cut rates, even though this will hurt savers and help borrowers. But because this has political consequences, politicians may not cut rates as required, and so damage the economy as a whole.

    4. In a recession WITH NO OTHER UNBALANCES (serious ones that is) you need to cut rates. If you have unbalances galore like now (allthough not completely similar in all countries) you still end up in a political game.
      Which unbalance you attack first for instance.
      Here (and in Germany at the moment) you have either a RRE bubble (and a pensiongap) or higher interest rates. Probably best if via legislation re collateral this was adressed but it wasnot, anyway hard in Germany to do as they already have one of the strictest rules in that respect.

      This priority setting is distributive as well.

      When you go away from a clear cut situation it simply always becomes political.

      Not unlogical that you would give the economy the highest priority but it simply becomes a much tougher sell.
      On the other hand solving one bubble with the next one is a thing that has to stop. Monetary (and fiscal) policy should reduce the number of crisis and the magnitude thereof and not make them more frequent and larger.

    5. "Indeed, I think that it is partly because optimal monetary policy has distributional consequences that delegation is useful, otherwise partisan politicians start deviating from the optimum to please particular groups."

      Eh? As opposed to the delegates, whose "optimal monetary policy" you agree have "distributional results" which, by definition, "please particular groups"?

  4. 1. Watch how the present thing plays out. If we get a follow up crisis CBs and banks will very likely get the blame (and traditional politics will get a next hit), simply game over no more independent agencies stuff.

    2. RE is probably more important than the average size stimulus. As I see it likely the main reason why Holland got a hit and the main reason why Belgium/France are still going to get one. You have to keep prices somewhere in line otherwise you are either in no consumption lad on one side and bubblebursting in time on the other.
    Simply have to avoid bubbles there. And Skandinavia has low interest because of the worldwide mess and its economies didnot take a big hit. So a bubble developping is very likely otherwise.
    Personally I do not see much merit in solving (better trying to) the last crisis (exploding RE bubble in the US), with creating a new one. And in that process f up the sov debt levels and CBs BS with no realistic easy way out.

    3. Montesquieu revisited. CB/money stuff is one of the main functions of government. It is not included in the trias thing (probably as it wasnot relevant in the 17-18th century. Judiciary can be kept out as it is basically a technical thing (largely not making political choices). When that gets more political like in the US problems arise as well btw.
    CBing has to be put in that scheme . One way or another. The combination political choices and not being elected in that respect is not a logical one. 2 Braincells voting isnot either. Have to find some equilibrium there. Spreading functions. Which opens however coordination issues.
    Anyway these checks and balances stuff also doesnot work well with traditional government (basically too much power has moved to the executive). Looks in for a resettle imho (but will likely take years probably decades). Hard to see CBing etc will be properly arranged other than ad hoc (as now the socalled banking (non-) union in the EZ). And with no 'structure' in the set up (so likley a recipe for future disaster).

    1. To give unelected institutions credibility they better be predictable.
      Spreading functions and make the individual functions/mandates basically technical is an idea.

      Another point is that what most macro folks seem to miss, that wiping your backside with your mandate (as basically the ECB is doing) hardly will do your fuctioning in general much good longer term.
      Make better laws and if they donot work change them, not use them as toilet paper like we see now too often. ECB has this way lost it credibility with large parts of society, not sustainable that way and making itself in advance the scapegoat if things end badly (big strategic gamble).

      You get a coordinationproblem very likely.
      All should be focussed by not having unbalances develop. So the firebrigade should only have to fight one fire at the time.
      I am doubtful (missing the RE bubble and now the Equity one hardly makes one optimistic it should work) if that can be achieved.
      If so the worst that can happen is that a structural crisis is mistaken for a cyclical one (or the structural part underestimated). But that is not a huge problem as by far the most dips are cyclical (providing you donot push the repeat button as standard solution like now). Furthermore as we see now there has to be a political platform created for structural reforms and that takes time and very likley requires a crisis anyway.

  5. I think it would have been pretty hard for the Riksbank to have acted in any other way than it has actually done.

    Why is this?

    Sweden does actually have a real estate bubble of pretty epic proportions, and when Swedes end up "underwater" with their mortgages and are foreclosed upon, they can't hand the key back to the bank and get the rid of the loan. Instead, the loan stays with them until every single krona has been payed back, or the debtor dies, whichever happens first.

    According to a study by Finansinspektionen, a 30% fall in real estate prices would put 50% of all Swedish homeowners underwater. About 75% of all Swedish households are homeowners. The situation becomes even worse when you consider that most Swedes have variable rate mortgages.

    Furthermore, the government has flat-out refused to do anything about the problem. Eliminating rent-control is a third rail of Swedish politics. Eliminating the 30% tax deduction on on all mortgage interest expenses is an even higher voltage third rail. Introducing a regulation that forces debtors to amortize their mortgages over time is incredibly controversial. More than half of all Stockholm mortgage holders do no amortize their loans at all. Suddenly people would have to cut back consumption to pay down debt, rather than using their houses as ATM's to finance new Volvos and vacations to Thailand.

    Finally, the Riksbank has very explicitly announced that it would love to cut interest rates right away if the government would implement some of the reforms I mentionded above. The answer from the government has been a deafening silence.

    1. Thank you for this. Your remarks on the lack of cooperation from government is interesting. Does the same apply to macroprudential regulation? The speech I linked to talked about developing these tools, but I did not understand why the process was taking so long.

    2. My take is that there is disagreement between the Riksbank and the ministry of finance about the pace of reform. 2014 is an election year. Anders Borg wants to put a brake to the house prices, not cause a collapse. For now the plan seems to be to increase the risk weights of mortgages stepwise up to 35%. Due to very low historical delinquency rates (in part due to the difficulty for distressed debtors to get relief, as mentioned by Arvid) these weights were very low. They are now raised to 15% but more is needed if it is to have any real effect on mortgage rates and prices.

      Of the options listed by Arvid I believe some kind of forced amortisation would be the most likely policy tool to be implemented. However, any implementation would need to be gradual, with the Riksbank matching step with reductions of the interest rate, in order to avoid big corrections in the housing market.

      Other policy options suggested by IMF is to reintroduce the property tax and to stream-line the planning laws to facilitate more residential construction. I do not think the present government will re-introduce the property tax (one of the signature policy changes of their first mandate), but easing some of the restraints on building might be a starter. It will however take long time until such changes trickle through into house prices.

      The property tax reform was revenue neutral and introduced a limit on the deductions for deferred profits and a kind of rent tax on the deferrals. I would say these changes are now starting to have an impact on the higher end of the market, as any sale will inevitably lead to profits bigger than the max deduction and a 22% tax payment. This part is missed by most foreign observers, including the IMF.

      The government is right now engaging in deficit spending to counterbalance the tight monetary policy. Perhaps ironically, the historical budget surpluses of the government has forced pension funds and other investors to switch to resold mortgage obligations, making an abundance of cheap money available to the housing market.

    3. Simon, the Swedish Financial Supervisory Authority (Finansinspektionen) has been very slow and reluctant when it comes to measures to counter growing household debt and housing prices. Representatives from Finansinspektionen have also been active deniers of any Swedish housing bubble. The reasons for this are open to speculation...

  6. Very nice article. I do agree that the Riksbank is endangering it's independence and overstepping it's mandate. Whether or not Sweden has a housing bubble or not is impossible to tell. There is an acute housing shortage in all towns with > 100 000 inhabitants, obviously this drives up prices.

    There is a turf war with the government recently handing bank regulation to Finansinspektionen. On can interpret the latest talk from Riksbank of "we will lower rates if you do this" as not accepting this.

  7. I went over this website and I believe you have a lot of wonderful information, saved to my bookmarks
    Financial Beratung Schweiz

  8. I have gone through this site and it is really interesting, for more details follow our site:-The new construction homes Valencia at this time mainly federations the importance of preserving bed-sitters at that juncture it can variation your brand to have a creators California home builder discounts disposition. The sector is communicated in boundless between by way of all new home builder CA expectation in observance on the other hand not so moral for additional California home search for the difference or different the happing personal property.


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.