Intro

I’ve been asked to talk about the ‘mistakes made’ with regards to the Global Financial Crisis, and the ‘financial architecture.  Let us start off by not assuming a mistake per se, but in asking what type of narrative describes what we experienced – from that we can try to describe what can be seen as a mistake.  Furthermore, the term “architecture”, or “infrastructure”, also presupposes something about the nature of financial institutions – and directly implies it is something that needs to be built, making it sound as if we need a central organisation to do it.  While there might be some truth in that, it is not where I’d prefer to start with any discussion.

Focusing on supranational, undemocratically determined, institutions for “solutions” to a policy problem misses the point – yes the world is integrated and national level institutions need to coordinate more, but focus needs to be on national policy when discussing mistakes.  As Dani Rodrik says, it is up to nations to keep their house in order, not other countries.  If anything the crisis in Europe shows the shortcoming of supranational regulation, given the true democratic mandate is a national one.

Even after a crisis the key thing we need to ask is what narratives fit our data, so we can build an understanding of what the impact of (national level) regulatory policy could be.  For an example of how this matters, say that if a financial crisis is like an earthquake – in that case we may believe that policy actions that help to insure people against the costs are appropriate.  Instead, assume that a financial crisis was due to the risky choice of individuals and organisations – then we would want policy that makes these people bear the costs of their choice.

The “truth” is between these, so let’s talk about these narratives.

The financial system

The GFC should not have changed our view of broad view of what financial institutions are.

The purpose of the financial system is as a market – as place where individuals and groups who want to save match up with individual and groups who want to borrow.  The institutions in the middle are financial intermediaries – and their role can be viewed in a number of ways:

1. By screening the quality of borrowers, and doing so with the economies of scale of a large institution, they provide a cost effective service for those who want to save.
2. Due to their scale, financial institutions are able to diversify risk – which individual depositors may not be able to do if they lent directly.
3. Maturity transformation:  Financial institutions, due to their size, are able to “lend for the long term” and “borrow for the short term” as long as depositors have faith in the institution – this ‘maturity mismatch’ creates value for both lenders and borrowers.

There are broader points about banks offering a ‘risk-free rate of return’, determining the money supply, inflation-protection of savings, and offering financial services – these are important issues when thinking about regulation, but are a bit beside the point when we are focused on the GFC directly!

The question we are focused on is around the “mistakes” made by policy during the Global Financial Crisis.  We can split mistakes into two categories:

1. Mistakes in terms of policy during the crisis.
2. Mistakes in terms of regulation and policy in the lead up to the crisis.

This separation is very important, as it helps to give us a perspective on the way policy and institutions can be changed to reduce the chance of a similar crisis – and also helps us to think about the cost of these regulatory changes!

The catalyst of the crisis

When thinking about the crisis kicking off, a lot is made of subprime mortgages, and non-recourse mortgages (so the debt is against the collateral in the house, not pinned to the individual borrowing).  Although housing is an important asset class (and the structure of mortgages did influence the way the market adjusted), it is more useful to think about these events in more general terms if we want to consider the financial infrastructure more generally.

• Lehman Brothers and the failure of AIG
  • Crisis had kicked off in mid-2007 with the housing market slowing in the US and BNP Paribus (in France) running into a liquidity crisis in August.
  • Northern Rock in February 2008 and the Bear Sterns event in March 2008 – still debate about whether the bailout exacerbated risk taking in the lead up to September, or whether it did prevent a crisis as of March.
  • Failure of LB after Barclays refused to buy at the last minute in September 2008.  Led to the near failure of AIG, an insurance company that had essentially been insuring a sizable chunk of the shadow banking system.
  • “Oil price shock”.  The steep run up in oil prices during 2007/08 was also a major, and often underplayed, cause of the global recession in late 2008/early 2009.  The magnitude of this shock was underplayed at the time, and was part of the reason why the slowdown heading into during 2010/11 was also more intense than anticipated (oil prices were rebounding at this time).

The response

After it became apparent that the failure of Lehman Brothers had done irreparable damage to AIG, faith in the entire financial system collapsed.  The rapid nature of the response by global economic authorities is captured in a book called “the alchemists” by Neil Irwin.  This gives a good run down of what happened through the narrative of the US, UK, and European central bankers at the time.

A combination of monetary policy easing, currency-swap agreements between central banks, significant new liquidity facilities (eg accepting lower quality collateral when giving loans), and direct purchases of some institutions (where shareholders value was essentially wiped out – but bondholders retained their claims) were all implemented in the wake of the Lehman Brother’s crisis.  Together, these policies helped to stabilise the financial system.

Fragile by design or nature?  Why not both.

The response to the crisis was largely appropriate.  Actions by the Federal Reserve and US Treasury helped to stem what was a “bank run” in the shadow banking system.

Shadow banking system:  “Shadow banking, as usually defined, comprises a diverse set of institutions and markets that, collectively, carry out traditional banking functions–but do so outside, or in ways only loosely linked to, the traditional system of regulated depository institutions – Bernanke”

The idea of a bank run in the shadow banking system was popularised by Gary Gorton.  He noted that one of the key reasons the crisis had such a broad impact on the financial market is because of the nature of shadow banking, and the way the line was blurred between commercial and shadow banking – namely that commercial banks would use the shadow banking sector for a significant amount of their interbank lending and borrowing activity.  One of his views was that the shadow banking system evolved to avoid regulation (rather than being the result of lax regulation per se) – which gives him the conclusion that we simply need to extend regulation to capture the shadow banking system.

Of course, this is far from easy – trying to regulate a specific industry is like trying to squeeze a lemon with your hand and keep it contained, the harder you squeeze the more bits leave your hand completely.

When faced with a bank run in a situation where large parts of the financial system are inherently “solvent” but facing problems of “liquidity” a government can improve outcomes by saying they will not let the organisation fail.

Illiquidity:  When an institution would, if holding its assets to maturity/selling at a ‘normal’ point of time, would be able to cover its liabilities.  However, during a crisis the institution is unable to sell its assets at a sufficiently high value to meet its liabilities.

Insolvent:  When an institutions liabilities would exceed the value of their assets in normal economic circumstances.

A bank run leads to a coordination failure for solvent firms with liquidity issues – if people were not pulling their funds out of a solvent financial institution, then it is in investors interest not to take out their money.  But if everyone is taking out their funds, the institution is forced to sell assets at “firesale” prices, making it in the investors interest to pull out as soon as possible!  This illustrates that the importance of the bank run is closely related to the idea of maturity transformation (or maturity mismatch) discussed above!

Note:  It can be hard to separate the two – given that insolvency may only become clear during a crisis!  When Bagehot came up with this in in the 19^th century he was clear that insolvent institutions should be allowed to fail, but modern regulators realise judging who is “insolvent” is unclear.

Mistakes leading up to the crisis?

However, we can’t just think of these issues at a point in time – the existence of a guarantee in the case of failure changes how we all view institutions.  If such a bailout is expected, depositors with financial institutions are willing to accept a lower rate of return – as a result, the entire financial system ends up taking on more risk.

This is a version of “moral hazard” and stems from the financial system (and depositors) being implicitly or explicitly insured by central government.

While there is this policy view of risk, there is also the more general ideas of “externalities” and “systemic risk”.

The key concern here is, even if firms were appropriately taking into account risk, they do not take into account how their failure could impact upon other institutions.  In this case, there are two issues of note:

1. A type of externality where the risk taking of one financial institution negatively impacts on other institutions.
2. A “cascading failure” in financial institutions due to the interdependence of firms – implying that in the face of a “systemic event” the failure of one institution could lead to the failure of the entire financial system.  This is similar to the idea of bank runs – and again a clear difference between “liquidity” and “solvency” needs to be taken into account.

However, it is clear that the policy and the market risks play off each other – the general unwillingness of policy to let institutions fail incentivises risky lending which makes the financial system more fragile, both in terms of having “too big to fail” (TBTF) firms and “too interdependent to fail” (TITF) firms. [Note this is very different to the Great Depression, where it was the fact firms were too small, and so were all treated the same independent of actual quality, that helped drive mass bank runs – these issues are often unclear!].

We can only say there is a “policy mistake” if we are clear about the goals of policy.  There are some key points that come from our narrative:

1. Some of the advantages of financial intermediation also increase fragility – implying there is, along some margin, a trade-off between efficiency and stability (eg maturity mismatch)
2. Although we have split the failures into three categories (regulatory, externality, systemic event) the three are intrinsically related and can feed off each other.
3. We need to be clear about the difference between insolvency and illiquid institutions – financial institutions should be able to fail, and if an entire financial system was “insolvent” the decision about what to do is actually a conversation of who should bear the burden for the loss.
4. More generally our “pre-emptive” solutions depend strongly on what we believe the cause of the crisis is – if financial firms require regulation, regulation should be tied strongly to a perceived failure.
   1. In this context, clear capital ratios for banks have attractive properties – current policy about matching maturities, LVR limits, core funding ratio, risk-weighted capital ratios all justified in this light.  However, questions about how much the central bank should be “determining risk tolerance” for private institutions – something that makes the Admati & Hellwig (in The Bankers New Clothes) suggestion of a general capital/equity ratio more attractive, as it is based largely on the view that it is the “favouring of debt financing to equity financing of lending” driving a lot of the issues down the line (TBTF, risky behaviour).  Eg firms have the incentive to become TBTF solely because of the implicit subsidy on debt financing, through the lender of last resort and tax status of this funding.
   2. However, capital requirements will still lead to excessive risk taking – although they may be helpful they are not the silver bullet that some may claim.  A recent paper by Josh Hendrickson (http://object.cato.org/sites/cato.org/files/serials/files/cato-journal/2014/2/v34n1-7.pdf) points out that one of the key issues is to ensure that bank shareholders face losses that are commensurate with the risks they took on in the face of a negative outcome – his suggestion is to replace limited liability with contingent liability.
   3. Note that Josh’s suggestion would help to deal with the typical moral hazard issue, but not necessarily with the fundamental issues of systemic risk – and expectations of a policy response in the face of a systemic event (which then leads us back to moral hazard).  It is along these lines that the RBNZ is trying to clarify rules with the Open Bank Resolution.

An inability to get lenders and borrowers to see eye-to-eye:  The case of Europe

However, the crisis did not end up finishing at this point.  After the global economy recovered sharply during the second half of 2009, events in Europe took a life of their own.  There are two clear narratives that given for the fact that global economy continued to struggle in the intervening five years:

1. A failure of monetary/government policy.  This includes the delays to implementing forward guidance, the fiscal tightening in a number of countries, and the issues of “coordination” with regards to policy between countries.
2. Policy and regulatory failure in Europe that kept acting as surprise “negative shocks” for the rest of the world.

The European Crisis was in many ways a different beast, as the concern was about whether individual nation’s government would pay bondholders.  An unwillingness to define whether the burden of “bad government debt” would fall upon bondholders, bank shareholders, or taxpayers created a situation of uncertainty – and led to a seizing up in financial markets.

Disappointingly, the events in Europe didn’t teach us particularly more about the Global Financial Crisis.  Instead, the primary additional (and extremely costly) lesson of the European Crisis was simply that the institutions involved in European governance are weaker than we had realised.

Cheers Statistics NZ.

This supports the type of story we have seen from the employment rate data, and our suggestion here that the 87-92 period was incredibly ugly for NZ.  As a result, even though in GDP terms the current slowdown looks like one of the worst NZ has experienced (though not so much in RGNDI terms – given the lift in NZ’s terms of trade) overall employment and unemployment outcomes have thankfully been nowhere near as bad as in the early 1990s.  I’m looking forward to all the analysis that will appear sifting through the data and working out why this is the case

Given how well his article held up, I was tempted to see what embarrassing things I said at the time – given that the magnitude of the deterioration was definitely worse than I could tell in real time.  Luckily, my posts are simply descriptive things pointing out some of the factors we need to keep an eye on (here on the day, here a few days later) – my view of the situation had shifted significantly since the failure of Bear Sterns in March (where I comment about moral hazard in comments).

When I think back to those times my way of compartmentalizing the issues was the same, but my assumption about a central banks reaction function was very different – I had thought that central banks would be “conservative” in terms of bailing out financial institutions more readily than they would optimally need to, as a result moral hazard appeared to be a more timely issue than the actual collapse of a financial institution which they would ‘never allow’.  My expectations, and the expectations of many people in financial markets, took a bit of a shock when Barclay’s dropped out of the purchase and the Fed allowed Lehman’s to simply declare bankruptcy.

The OBR, macroprudential policy, comments about asset bubbles – these are all central banks feeling out ways they can improve policy in the financial stability space.  Prior to the crisis, there for analysts there was an assumption that either there were policies in place to avoid a crisis (a minority) or that central banks could quickly and sufficiently respond to a crisis (a majority, eg here although I am thinking also in terms of consistent expectations in financial markets) – and it is arguments about “why this failed” that is driving a lot of discussion forward.  And this is good.

The regulatory reforms that have since been pushed through at Basel read as an extended mea culpa by central bankers for getting things so grievously wrong before the financial crisis. But regulators and bankers were not alone in making misjudgments. When economies are doing well there are powerful political pressures not to rock the boat. With inflation at bay central bankers could not appeal to their usual rationale for spoiling the party. The long period of economic and price stability over which they presided encouraged risk-taking. And as so often in the history of financial crashes, humble consumers also joined in the collective delusion that lasting prosperity could be built on ever-bigger piles of debt.

A lot of what they say in the article is true, although I will be honest that I don’t fully agree with their description of the lead up to the crisis … and as a result, will probably have some differences of opinion in terms of what I view as appropriate lessons.  However, I will leave all this for another time [Note:  Many of these things involve markets – but then many of the explanations involve only looking at “one side”.  This is common, but always a touch disingenuous IMO]

Feel free to discuss the GFC, and the description of the causes by the Economist, in comments below!

But debt doesn’t make workers want to work less, it makes them want to consume less.  There is a difference.  We need economists to look through these framing effects, and see that the standard model that demand shocks cause high unemployment worked fine; it’s our policymakers who failed us.

There is a huge difference.  Economists say GDP = C + I + G + X – M – but in macro you can’t just take these individual factors and talk about their impact on GDP, they are all massively endogenous.  In other words, this decomposition can be misleading – and gets filled with “fallacy of composition” issues when people try to use it for policy!

“GDP” is produced by factors of production, it is output using labour and other factors.  People are worried about high unemployment, and underutilised resources, this is exactly what Sumner is talking about when discussing active monetary policy of “demand management”.

The obsession with including “finance” into the cycle has to be viewed carefully.  In so far as it improves forecasts, and thereby improves the ability to “target a path” for demand, and improves our understanding of “what could or will happen” it is very very useful.  But it doesn’t actually change the fundamental relationship that monetary authorities should be focused on when setting monetary conditions!

Now this is where I get concerned with Borio’s stuff.  There is a lot of very good work in there (I agree with a lot of the discussion of the financial industry – and using financial indicators to “add information” about the output gap), but he is obsessed with selling his work “output gap that is adjusted for finance” or a “finance neutral output gap”.  This framing doesn’t make any sense to me – and in my reading of his papers he has never provided a full methodological reason why the output gap measure should be sold in this way.  He attempts it post-hoc justifications of why finance impacts on deviations from potential bullet two of this Vox Eu article – but this is still an uncompelling base.  When it comes to monetary policy the actual counterfactual we are interested in is in terms of unemployment and inflation – other factors matter (in terms of monetary policy) only INSOFAR as they influence these!

Macroprudential policy yadda yadda yadda are structural policies about “real economy” issues – monetary policy is set GIVEN these.  But having monetary authorities focus on this instead of actual monetary conditions is missing the wood for the trees.  Financial market information is useful to help inform us of where we are, and where we are going, BUT can we actually focus on the first order issue of the purpose of “active monetary policy” which are medium-long term inflation outcomes and short term unemployment variation (read the short-term as the outlook for 18-24 months out, I realise this term can be vague otherwise!).  If the financial market information helps the central bank set policy to deliver this (which it will) use the hell out of it – but don’t then lose sight of your actual purpose and start trying to value companies and assets for the market at large.  If you want to do that monetary institutions, give up your contract with government and get a job as a financial adviser

When I read this sort of stuff all I see is monetary theorists saying “ahhh, NGDP growth was weaker than authorities were committed to delivering, here are some excuses”.  But as soon as they give up the role they were given by government (deal with inflation and short term fluctuations in the UR) they leave themselves increasingly open to arbitrary demands and politicisation.

After ignoring prudential standards and arbitrarily trying to deal with it with monetary policy, we stumbled into the Great Depression.  Pressure post Great Depression and war, in time led to the hyperinflation and stagflation through the 1970/80s, which led to a recognition that policy needed to narrow.  Which led to completely ignoring macroprudential rules (which is not good either) which then led to the 2008/09 crisis. Which led to …

8) Can we realistically solve the “too big to fail” problem?

We have to solve it. If we can’t, then nationalize these behemoths and pay the people who run them the same wages as everyone else who work for the government.

Fair call, economists, like most people, despise “socialism for the rich”.  And honestly, if public institutions are always going to backstop them then nationalisation makes a lot of sense – and is how we felt here during some of the financial firm bailouts.

I can’t see the government letting banks fail, which is why I’ve been pro-deposit insurance recently (here, here).  But even down that road, have a deposit levy.  Also ask why we are going down that road, is it because we want a “risk free” rate of return for mum and dad investors no matter the cost … if that is the social preference Bair’s suggestion of nationalisation starts to make a lot more sense right

I’d note something that doesn’t come from these posts – some other people are simultaneously saying QE does nothing to increase NGDP, and that it creates asset bubbles.  This argument is far from clear to me, if QE is doing nothing because it is merely a swap of assets with the same yield with no macro impact then why do asset prices change?  Isn’t the increase in asset prices partially due to a lift in expected NGDP growth.  “Bubble” isn’t an answer to everything – we sort of need to understand where it is coming from.

I am sort of hoping some people may stumble here to tell me what I’m missing

Update:  AAMC on twitter reminded me about this post, which I’d seen as well.  The Cochrane post and comments are, to my mind, a discussion on the same sort of issues – and the Sumner post is a reminder that we have to think about the monetary policy rule and the expectations channel, and base money is special because it is a medium of account (I’m a bit nervous here, as I suspect T-bills are also seen as a medium of account in a large number of contexts, which takes us back to the Cochrane post).

All in all, the debate about QE as a mechanism is important.  Thank goodness we are not in this situation in NZ.

There is strong evidence that house prices and consumption are synchronised. There is, however, disagreement over the causes of this link. This study examines if there is a wealth effect of house prices on consumption. Using a household-level panel data set with information about house ownership, income, wealth and demographics for a large sample of the Danish population in the period 1987–96, we model the dependence of the growth rate of total household expenditure with unanticipated innovations to house prices. Controlling for factors related to competing explanations, we find little evidence of a housing wealth effect.

• The initial post.
• Tyler Cowen’s thoughts.
• The R-R reply. (Here)
• A post suggesting that it mattered for policy settings.

All very nice.  Depending on the message people were trying to sell they either said the result was meaningless, or central, so I don’t think this makes any actual difference.  Honestly, without clear causal drivers there just were not good evidence based claims for actual policy adjustments – a lot of people were actually just saying we need to do X based on their preconceptions.  And they found this correlation either something that supports that (somehow) or something they need to rule out.

Over the last few years I have seen authors, at different times, use R-R as central to their argument on one thing, and then dismiss it for arguments regarding other issues (I’m not going to name names).  For example, you can’t use this result to say we need more savings policy because the stock of debt is to high, then complain that the study is flawed when you want more government borrowing … just focus on the actual core elements of your frikken argument instead!

My problem with the result isn’t the excel errors, or anything R-R appear to have said – it is the way it has been used as an inconsistent marketing tool by people for selling their own unrelated ideological policies.  I’m just hoping that this shuts that up.

As a side note, here are my feelings on twitter:

People who think the R&R result caused austerity overestimate the impact evidence has on government policy.

Now when I’ve described the crisis to people I’ve stuck to the inherent instability line, the trilogy of articles (*,*,*) I did on Rates Blog was supposed to give that impression to people – with the third article pointing out that moral hazard exists as a more long-term point.  According to this, the Fed and then the ECB didn’t do enough to stop a bank run due to “too much” weight on moral hazard – and this drove a deep crisis.

However, the induced fragility hypothesis is also compelling (note they could both have happened – but where you put the weight determines what policy lessons you learn).  According to this, it wasn’t until the policy actions of late 2007 and then the bail out of Bear Sterns then the moral hazard became compelling.  However, once financial institutions saw they would be bailed out, they immediately took on lots of risk – making the system a lot more fragile when the Fed finally decided not to bail out Lehman Brothers.

There is evidence for this, Lehman Brothers took on a lot of debt – highly risky debt – following the collapse of Bear Sterns.  I remember reading the paper that had this, a paper that actually said it was the bailout of Bear Sterns that made the crisis worse, but I have forgotten where it is … I’ll link when I find it.

If the spectre of moral hazard fed into the fragility of the finanical system that quickly, the justification for bailouts becomes a lot weaker – if we accept significant inherent instability in the financial sector, then regulation with a lender of last resort becomes more acceptable.  Evidence helping to determine what weights to put on these explanations will be very useful for designing regulation (or the lack of) in the post-GFC world.

Induced fragility as a medium term concept is a central part of how most economists see this crisis.  However, it is an open question about whether the bailout of Bear Sterns (and the earlier TAF) created “induced fragility” that made the collapse much worse.  More research on this issue will be pretty interesting, and will help to inform what sort of trade-off we are facing with policy.