David Beckworth wrote a good post on the secular stagnation hypothesis, highlighting the problem with the interpretation of the natural rate by secular stagnation theorists. This is Beckworth:
First, real interest rates adjusted for the risk premium have not been in a secular decline. Everyone from Larry Summers to Paul Krugman to Olivier Blanchard ignore this point in the book. They all claim that real interest rates have been trending down for decades. The editors of the book, Coen Teulings and Richard Baldwin, even claim that this development is the ‘prima facie’ evidence for secular stagnation. What they are doing wrong is only subtracting expected inflation from the observed nominal interest rate. They also need to subtract the risk premium to get the natural interest rate, the interest rate at the heart of the story. For it is the natural interest rate that is affected by expected growth of technology and the labor force.
He is right. But I believe there is more to the story.
This is Beckworth’s chart highlighting the apparent secular decline in interest rates as believed by secular stagnation adherents. I added to it Basel 1, Basel 2 and Basel 3 introductions (red lines) and discussions (orange area) (unlike in Europe, only a few US banks had implemented Basel 2 when the crisis started):
What is striking is the fact that Treasury yields started declining exactly when Basel 1 was being discussed and implemented. There is a clear reason for this: Basel introduced risk-weighted assets, and government securities were awarded a 0% risk-weight, meaning banks could purchase them and hold no capital against them. As I have described before, banks were then incentivised to pile in such assets to maximise RoE, as a quick risk-adjusted return on capital calculation demonstrates. Basel 1 also introduced risk-weights for other asset classes such as business and mortgage lending. Structural changes in lending* and government securities markets occurred directly post-Basel. I believe this is no coincidence.
Take a look at the following chart. Until the 1980s, the volume of US government securities on US banks’ balance sheet was pretty much constant. Everything changed from the 1980s onward:
As a share of banks’ total assets (see chart below), US government securities literally spiked after Basel 1 was introduced, only to decline (as a share of total assets) as banks started piling in other assets that benefited from generous capital treatments such as securitisations and insured mortgages (though this doesn’t mean demand faded as banks balance sheets grew quickly over the period, just that demand for other assets was even stronger). Post-crisis, Basel 3 renewed the demand for US sovereign debt as it 1. modified the capital treatment of some previously lowly-weighted assets and 2. introduced minimum liquidity requirements (LCR) as well as margins and collateral requirements that require the use of high-quality liquid assets such as Treasuries.
We identify the same pattern as a share of total securities on US banks’ balance sheet**:
The demand for Treasuries also boomed throughout the financial sector due to those margin/collateral/liquidity requirements that apply (mostly post-Basel 3), not only to commercial banks, but also to broker dealers and investment managers:
All those regulations evidently artificially increase demand for US government-linked securities, pushing their yields down.
But I guess you’re going to tell me that Beckworth’s adjusted risk-free Treasury yield was actually stable over the whole period (chart below). Actually, unlike what Beckworth claims, it does look like there is a slight decline since the end of the 1980s. Moreover, ups and downs almost exactly coincide with banks decreasing/increasing their relative holdings of Treasuries (see above, third chart). Finally, there is also another option: that the natural (risk-free) rate of interest’s trajectory was in fact upward, which would be hidden by financial regulations’ artificially-created demand…
However, this analysis is incomplete: it does not account for foreign banks’ demand for Treasuries, which is also a widely-held asset as part of banks’ liquidity buffers (and in particular when their own sovereign fails…). I unfortunately don’t have access to such data.
In the end, the use of Treasury yield (even adjusted) as an estimate of the natural rate of interest is unreliable given the numerous microeconomic variables that distort its level.
* See this chart from one of my previous posts:
** Post-WW2’s very high figures (close to 100%) reflect the low level of corporate securities issued following the Great Depression and WW2, the high issuance volume of US sovereign debt to fund the war, as well as restrictions on banks’ securities holding to facilitate such wartime issuances.
I recently mentioned David Beckworth’s excellent new paper on inflation targeting, which, according to him, promotes financial instability by inadequately responding to supply shocks. Like free bankers and some other economists, Beckworth understands the effects of productivity growth on prices and the distorting economic effects of inflation targeting in a period of productivity improvements.
Nevertheless, I am a left a little surprised by a few of his claims (on his blog), some of which seem to be in contradiction with his paper: according to him, current US output gap demonstrates that the nominal natural rate of interest has been negative for a while. Consequently, the current Fed rate isn’t too low and raising it would be premature.
While I believe that the real natural interest rate (in terms of money) is very unlikely to ever be negative (though some dispute this), it is theoretically and empirically unclear whether or not the nominal natural rate could fall in negative territory, especially for such a long time.
Beckworth uses a measure of US output gap calculated by the CBO and derived from their potential GDP estimate. This is where I become very sceptical. GDP itself is already subject to calculation errors and multiple revisions. Furthermore, there are so many variables and methodologies involved in calculating ‘potential’ GDP, that any output gap estimate takes the risk of being meaningless due to extreme inaccuracy, if not completely flawed or misleading.
This is the US potential GDP, as estimated by the CBO:
Wait a minute. For most of the economic bubble of the 2000s, the US was below potential? This estimate seems to believe that credit-fuelled pre-crisis years were merely in line with ‘potential’. This is hardly believable, and this reminds me of the justification used by many Keynesian economists: we should have used more fiscal stimulus as we are below ‘trend’ (‘trend’ being calculated from 2007 of course, as if the bubble years had never happened). Does this also mean that the natural rate of interest has been negative or close to zero since 2001? This seems to contradict Beckworth’s own inflation targeting article, in which he says that the Fed rate was likely too low during the period.
Let’s have a look at a few examples of the wide range of potential GDP estimates (and hence output gap) that are available out there*. The Economic Report of the President estimates potential GDP as even higher than the CBO’s (source: Morgan Stanley):
The Fed of San Francisco, on the other hand, estimated very different output gap variations. According to some measures, the US is currently… above potential:
Some of the methodologies used to calculate some of those estimates might well be inaccurate, or simply wrong. Still, this clearly shows how hard it is to determine potential GDP and thus the output gap. Any conclusion or recommendation based on such dataset seems to me to reflect conjectures more than evidences.
This is where we get to my point.
In his very good article, Beckworth brilliantly declares that:
the productivity gains will also create deflationary pressures that an inflation-targeting central bank will try to offset. To do that, the central bank will have to lower its target interest rate even though the natural interest rate is going up. Monetary authorities, therefore, will be pushing short-term interest rates below the stable, market-clearing level. To the extent this interest rate gap is expected to persist, long-term interest rates will also be pushed below their natural rate level. These developments mean firms will see an inordinately low cost of capital, investors will see great arbitrage opportunities, and households will be incentivized to take on more debt. This opens the door for unwarranted capital accumulation, excessive reaching for yield, too much leverage, soaring asset prices, and ultimately a buildup of financial imbalances. By trying to promote price stability, then, the central bank will be fostering financial instability.
Please see the bold part (my emphasis): isn’t it what we are currently experiencing? It looks to me that the current state of financial markets exactly reflects Beckworth’s description of a situation in which the central bank rate is below the natural rate. This is also what the BIS warned against, explicitely rejected by Beckworth on the basis of this CBO output gap estimate. (see also this recent FT report on bubbles forming in credit markets)
I am asking here how much trust we should place in some potentially very inaccurate estimates.
Perhaps the risk-aversion suppression and search for yield of the system is not apparent to everyone, including Beckworth, not helping him diagnose our current excesses. But, his ‘indicators that don’t show asset price froth’ are arguable: the risk premium between Baa-rated yields and Treasuries are ‘elevated’ due to QE pushing yields on Treasuries lower, and it doesn’t mean much that households still hold more liquid assets than in the financial boom years of 1990-2007.
At the end of the day, we should perhaps start relying on actual** – rather than estimated and potentially flawed – indicators for policy-making purposes (that is, as long as discretion is in place). Had US GDP been considered as above potential in the pre-crisis years and the Fed stance adapted as a result, the impact of the financial crisis might have been far less devastating. I agree with Beckworth: time to end inflation targeting.
* The IMF estimate also shows that the US was merely in line with potential as of 2007. Others are more ‘realistic’ but as the charts below demonstrate, estimates vary widely, along with confidence intervals (link, as well as this full report for tons of other output gap charts from the same authors):
** I understand and agree that ‘actual’ market and economic data can also be subject to interpretation. I believe, however, that the range of interpretations is narrower: these datasets represents more ‘crude’ or ‘hard’ data that haven’t been digested through multiple, potentially biased, statistical computations.
I often refer to the distortive effects of RWA on the housing and business/SME lending channels. What I don’t say that often is that Basel’s regulations have also other distortive effects, perhaps slightly less obvious at first sight.
Basel is highly likely to be partially responsible for sovereign states’ over-indebtedness, by artificially maintaining interest rates paid by governments below their ‘natural’ level.
How? Through one particular mechanism historically, that you probably start knowing quite well: risk-weighted assets (RWA). Basel 1 indeed applied a 0% risk-weight on OECD countries’ sovereign debt*, meaning banks could load up their balance sheet with such instruments without negatively impacting their regulatory capital ratios at all. Interest income earned on sovereign debt was thus almost ‘free’: banks were incentivised to accumulate them to maximise capital-efficiency and RoE.
This extra demand is likely to have had the effect of pushing interest rates down for a number of countries, whose governments found it therefore much easier to fund their electoral promises. In the end, the financial and economic crisis was triggered by the over-issuance of very specific types of debt: housing/mortgage, sovereign and some structured products. All those asset classes had one thing in common: a preferential capital treatment under Basel’s banking regulations.
Basel 2 introduced some granularity but fundamentally didn’t change anything. Basel 3 doesn’t really help either, although local and Basel regulators have recently announced possible alterations to this latest set of rules in order to force banks to apply risk-weights to sovereign bonds (one option is to introduce a floor). Some banks have already implemented such changes (which cost billions in extra capital requirements).
While those measures go in the right direction, Basel 3 has also introduced a regulatory tool that goes precisely the opposite way: the liquidity coverage ratio (LCR). The LCR requires banks to maintain a large enough liquidity buffer (made of highly-liquid and high quality assets) to cover a 30-day cash outflow. As you may have already guessed, eligible assets include mostly… government securities**.
Here again, Basel artificially elevates the demand for sovereign debt in order to comply with regulatory requirements, pushing yields down in the process. This has two consequences: 1. governments could find a lot easier to raise cash than in free market conditions (with all the perverse incentives this has on a democratic process unconstrained by economic reality) and 2. as sovereign yields are used as risk-free rate benchmarks in the valuation of all other asset classes, the fall in yield due to the artificially-increased demand could well play the role of a mini-QE, boosting asset prices across the board ceteris paribus.
We end up with a policy mix that contaminates both central banks’ monetary policies and domestic political debates. But, worst of all, it is a real malinvestment engine, which trades short-term financial solidity for long-term instability.
* Some non-OECD regions of the world also allow their domestic banks to use 0% risk-weight on domestic sovereign debt. For instance, many African countries are allowed to apply 0% weighing on the sovereign debt of their local governments despite the obvious credit risk it represents as well as its poor marketability (this is partly mitigated as this debt is often repoable at the regional central bank). Moreover, the same regulators prevent their domestic banks from investing their liquidity in Treasuries or European debt, with the obvious goal of benefiting those African states. Consequently, illiquid and risky sovereign bonds comprise most of those banks’ “liquidity” buffers, evidently not making those banking systems much safer…
** The LCR is partly responsible for the ‘shortage of safe assets’ story.
The BIS has got a hobby: to annoy other economists and central bankers. It’s a good thing. It published its annual report about two weeks ago, and the least we can say is that it didn’t please many.
Gavin Davies wrote a very good piece in the FT last week, summarising current opposite views: “Keynesian Yellen versus Wicksellian BIS”. What’s interesting is that Davies views the BIS as representing the ‘Wicksellian’ view of interest rates: that current interest rates are lower than their natural level (i.e. monetary policy is ‘loose’ or ‘easy’). On the other hand, Scott Sumner and Ryan Avent seem to precisely believe the opposite: that current rates are higher than their natural level and that the BIS is mistaken in believing that low nominal rates mean easy money. This is hard to reconcile both views.
Neither is the BIS particularly explicit. Why does it believe that interest rates are low? Because their headline nominal level is low? Because their real level is low? Or because its own natural rates estimates show that central banks’ rates are low?
It is hard to estimate the Wicksellian ‘natural rate’ of interest. Some people, such as Thomas Aubrey, attempt to estimate the natural rate using the marginal product of capital theory. There are many theories of the rate of interest. Fisher (described by Milton Friedman as America’s best ever economist), Bohm-Bawerk, and Mises would argue that the natural interest rate is defined by time preference (even though they differ on details), and Keynes liquidity preference. Some economists, such as Miles Kimball, currently argue that the natural rate of interest is negative. This view is hard to reconcile with any of the theories listed above. Fisher himself declared in The Rate of Interest that interest rates in money terms cannot be negative (they can in commodity terms).
Unfortunately, and as I have been witnessing for a while now, Wicksell is very often misinterpreted, even by senior economists. The latest example is Paul Krugman, evidently not a BIS fan. Apart from his misinterpretation of Wicksell (see below), he shot himself in the foot by declaring (my emphasis):
Now, what about the BIS? It is arguing that central banks have consistently kept rates too low for the past couple of decades. But this is not a statement about the Wicksellian natural rate. After all, inflation is lower now than it was 20 years ago.
Given that we indeed got two decades of asset bubbles and crashes, it looks to me that the BIS view was vindicated…
The second issue to note is that when the natural rate is higher than the money rate there is no necessary impact on the general price level. As the Swedish economist Bertie Ohlin pointed in the 1930s, excess liquidity created during a Wicksellian cumulative process can flow into financial assets instead of the real economy. Hence a Wicksellian cumulative process can have almost no discernible impact on the general price level as was seen during the 1920s in the US, the 1980s in Japan and more recently in the credit bubble between 2002-2007.
(Bob Murphy also wrote a very good post here on Krugman vs. Wicksell)
But there are other problematic issues. First, inflation (as defined by CPI/RPI/general increase in the price level) itself is hard to measure, and can be misleading. Second, as I highlighted in an earlier post, wealthy people, who are the ones who own most investible assets, experience higher inflation rates. In order to protect their wealth from declining through negative real returns (what Keynes called the ‘euthanasia of the rentiers’), they have to invest it in higher-yielding (and higher-risk) assets, causing bubbles is some asset classes (while expectations that central bank support to asset prices will remain and allow them to earn a free lunch, effectively suppressing risk-aversion).
If natural rates were negative – or at least very low – and the environment deflationary, it is unlikely that we would witness such hunt for yield: people care about real rates, not nominal ones (though in the short-run, money illusion can indeed prevail). But this is not only an ultra-rich problem: there are plenty of stories of less well-off savers complaining of reduced purchasing power.
Meanwhile, the rest of the population and overleveraged companies, supposedly helped by lower interest rates, seem not to deleverage much: overall debt levels either stagnate or even increase in most economies, as the BIS pointed out.
Banks also suffer from the combination of low rates* and higher regulatory requirements that continue to pressurise their bottom line, and have ceased to pass lower rates on to their customers.
In this context, the BIS seems to have a point: rates may well be too low. Current interest rate levels seem to only prevent the reallocation of capital towards more economically efficient uses, while struggling banks are not able to channel funds to productive companies.
Critics of the BIS point to their call to rise rates to counter inflation back in 2011. Inflation, as conventionally measured, indeed hasn’t stricken in many countries. In the UK and some other European countries though, complaints about quickly rising prices and falling purchasing power have been more than common (and I’m not even referring to house price inflation). This mismatch between aggregate inflation indicators and widespread perception is a big issue, which underlies financial risk-taking.
In the end, Keynes’ euthanasia of the rentiers only seem to prop up dying overleveraged businesses and promote asset bubbles (and financial instability) as those rentiers pile in the same asset classes. I side with the BIS in believing this is not a good and sustainable policy.
I also side with the BIS and with Mohamed El-Erian in believing in the poor forecasting ability of most central bankers, who seem to constantly display a dovish view of the economy, which apparently experiences never-ending ‘slack’, as well as the very uncertain effect of macro-prudential policies, which cannot and will not get in all the cracks. Nevertheless, many mainstream economists and economic publications seem to be overconfident in the effectiveness of macro-prudential policies (see The Economist here, Yellen here, Haldane here, who calls macropru policies “targeted lightning strikes”…).
While central banks’ rates should probably already have risen in several countries (and remain low in others, hence the absurdity of having a single monetary policy for the whole Eurozone), everybody should keep the BIS warnings in mind: after all, they were already warning us before the financial crisis, yet few people listened and many laughed at them.
Unfortunately, politicians and regulators have repeated some of the mistakes made during the Great Depression: they increased regulation of business and banking while the economy was struggling. I have many times referred to the concept of regulatory uncertainty, as well as the over-regulation that most businesses are now subject to (in the US at least, though this is also valid in most European countries). Businesses complaints have been increasing and The Economist reported on that issue last week.
In the meantime, while monetary policy has done (almost) everything it could to boost credit growth and to prevent the money supply from collapsing, harsher banking regulation has been telling banks to do the exact opposite: raise capital, deleverage, and don’t take too much risk.
In the end, monetary policy cannot fix those micro-level issues. It is time to admit that we do not live in the same microeconomic environment as before the crisis. What about cutting red tape to unleash growth rather than risk another financial crisis?
* Yes, for banks, rates are low, whichever way you look at them. Banks can simply not function by earning zero income on their interest-earning assets (loan book and securities portfolio).
PS: Noah Smith, another member of the anti-BIS crowd, has a nonsense ‘let’s keep interest rate low forever’-type article here: raising interest rates would lead to an asset price crash, so we should keep them low to have a crash later. Thanks Noah. The way he describes a speculative bubble is also wrong (my emphasis):
The theory of speculation tells us that bubbles form when people think they can find some greater fool to sell to. But when practically everyone is convinced that asset prices are relatively high, like now, it’s pretty obvious that there aren’t many greater fools out there.
Really? No, speculation involved buying as long as you believe you can get the right timing to exit the position. Even if everyone believed that asset prices were overvalued, as long as investors expect prices to continue to increase, speculation would continue: profits can still be made by exiting on time, even if you join the party late.
PPS: A particularly interesting chart from the BIS report was the one below:
It is interesting to see how coordinated financial cycles have become. Yet the BIS seems not to be able to figure out that its own work (i.e. Basel banking rules) could well be the common denominator of those cycles (which were rarely that synchronised in the past).
The IMF got the timing right. It published last week a new ‘Global Housing Watch‘, and warned that house prices were way above trend in a lot of different countries all around the world. The FT also reports here:
The world must act to contain the risk of another devastating housing crash, the International Monetary Fund warned on Wednesday, as it published new data showing house prices are well above their historical average in many countries.
As I said, perfect timing, as this announcement follows my previous post on the influence of Basel’s RWAs on mortgage lending.
As long as international banking authorities don’t get rid of this mechanism, we are likely to experience reoccuring housing bubbles with their devastating economic effects (hint for Piketty: and investors/speculators will have an easy life making capital gains).
PS: I am on holidays until the end of the week, so probably not many updates over the next few days.
(and vs. BoE’s FLS)
Following my previous post on the mechanics of ECB negative deposit rates, I wanted to back my claims about the likely poor effect of the central bank TLTRO measure on lending.
I argued that despite the cheap funds provided by the ECB to lend to corporate clients (particularly SMEs), Basel’s risk-weighted assets would stand in the way of the scheme as they keep distorting banks’ lending incentives (same is true regarding the BoE and the second version of its Funding for Lending Scheme).
I extracted all the most recent new business and mortgage lending rates from the central banks’ websites of several European countries. Unfortunately, business lending rates are most of the time aggregates of rates charged to large multinational companies, SMEs, and micro-enterprises. Only the Banque of France seemed to provide a breakdown. So most business lending rates below are slightly skewed downward (but not by much as you can see with the French case).
Using this dataset, I built a similar scenario to the one I described in my first RWAs and malinvestments post (as it turned out, I massively overestimated business lending rates in that post…). I wanted to find out what would be the most profitable option for a bank: business lending or mortgage lending, given RWA and capital constraints (banks target a 10% regulatory Tier 1 capital ratio). The results speak for themselves:
Despite the cheap ECB loans, and given a fixed amount of capital, banks are way more profitable raising funding from traditional sources to lend to households for house purchase purposes…
Admittedly, the exercise isn’t perfect. But the difference in net interest income and return on capital is so huge that tweaking it a little wouldn’t change much the results:
- I assume that all business lending is weighted 100%. In reality, apart from the French SME scenario, large corporates (often rated by rating agencies) benefit from lower RWA-density under a standardised method. This would actually raise the profitability of business lending through higher volume (and increased leverage), though not by much. Mortgages are weighted at 35% under that method.
- I assume that all banks use the standardised method to calculate RWAs. In reality, only small and medium-sized banks do. Large banks use the ‘internal rating based’ method, which allows them to risk-weight customers following their own internal models. Here again, most corporates can benefit from lower RWAs. But mortgages also do (RWA-density often decreases to the 10-15% range).
- Cross-selling is often higher with corporates, which desire to hedge and insure their financial or non-financial business positions. Corporates also use banks’ international payment solutions. This adds to revenues.
- Business lending is often less cost-intensive than retail lending. Retail lending indeed traditionally requires a large branch network, which is less the case when dealing with corporates (often grouped within regional corporate centres, though not always for tiny enterprises). However, retail banking is progressively moving online, providing opportunities to banks to cut costs and improve their profitability.
- The lower RWA-density on mortgages allows banks to increase lending volume and leverage. However, this also requires higher funding volumes. In turn, this should increase the rate paid on the marginal increase in funding, raising interest expense somewhat in the case of mortgages.
In the end, even if the adjustments described above reduce the profitability spread by 10 percentage points, the conclusion stands: banks are hugely incentivised to avoid business lending, facilitating misallocation of capital on a massive scale, in particular in a period of raising capital requirements… Moreover, banks also benefit from favourable RWAs for securitised products based on mortgages (CMBS, RMBS…), compounding the effects.
To tell you the truth, I wasn’t expecting such frightening results when I started writing that post… Please someone tell me that I made a mistake somewhere…
Central banks, regulators and politicians will find it hard to prop up business lending with regulations designed to prevent it.
The ECB has finally announced last week that it would be lowering its main refinancing rate from 0.25% to 0.15%, and that it would lower the rate it pays on its deposit facility from 0% to -1%. The ECB hopes to incentivise banks to take money out of that facility and lend it to customers*, providing a boost to the broad money supply and counteracting deflation risks.
The interest rate of the ECB deposit facility is supposed to help the central bank define a floor under which the overnight interbank lending rate (EONIA) should not go. The reason is that deposits at the ECB are supposedly risk-free (or at least less risky than placing the money anywhere else). Consequently, banks would never place money (i.e. excess reserves) at another bank/ investment (which involves credit risk) for a lower rate. When the deposit facility rate is high, banks are incentivised to reduce their interbank lending exposures and leave their money at the ECB (and vice versa). On the other hand, the main refinancing rate is supposed to represent an upper boundary to the interbank lending rates: theoretically, banks should not borrow from another bank at a higher rate than what it would pay at the ECB. In practice, this is not exactly true, as banks do their best to avoid the stigma associated with borrowing from the central bank.
Unfortunately, there is a fundamental microeconomic reason why banks cannot diminish their lending rate indefinitely. I have already described how banks are not able to transmit interest rates lower than a certain threshold to their customers due to the margin compression effect (see also here). Indeed, banks’ net interest income must be able to cover banks’ fixed operating costs for the bank to remain profitable from an accounting point of view. When rates drop below a certain level, banks have to reprice their loan book by increasing the spread above the central bank rate on the marginal loans they make, breaking the transmission mechanism of the lending channel of monetary policy. As we have already seen, in the UK, that threshold seems to be around 2%.
But the same thing seems to happen in other European countries. This is Ireland (enlarge the charts)**:
We can notice the margin compression effect on the first chart. The lending rate stops dropping despite the ECB rate falling as banks reprice their lending upward to re-establish profitability (evidenced from the second chart, where interest rates on new lending increase rather than decrease).
This is France:
This is Austria:
Clearly, what happens in the UK regarding margin compression also occurs in those countries. When the ECB rate dropped, outstanding floating rate lending rates also dropped (because floating rate lending is indexed on either the ECB base rate or on Euribor), causing pressure on revenues. As long as deposit rates could also fall (this varies a lot by country, as French banks never pay anything on demand deposits), the loss in interest income was offset by reduced interest expense. But once deposit rates reached 0 and couldn’t fall any lower, banks in those countries experienced margin compression and their net interest income started to suffer. Moreover, this happened exactly when loan impairment charges peaked because of increased credit risk.
Banks in emerging countries have constantly high non-performing loans ratios. But they still manage to remain (often highly) profitable by maintaining very high net interest income and margins. In Western Europe, banks saw their net profits all but disappear with rates dropping that low. As a result, it is likely that the new ECB rate cut won’t affect lending rates much…
Nonetheless, the ECB had been trying to revive, or encourage, interbank lending throughout most of the crisis, and had already lowered to its deposit rate to 0%. Nevertheless, banks maintained cash in those accounts. Why would a bank leave its money in an account that pays 0%? Because banks adjust those interest rates for risk. An ECB risk-adjusted 0% can be worth more than a risk-adjusted 4% interbank deposit at a zombie/illiquid/insolvent bank. However, the ECB is clearly not satisfied with the situation: it now wants banks to take their money out of the facility and lend it to the ‘real economy’.
How do the combination of low refi rate and negative ECB deposit rates impact banks? Let’s remember banks’ basic profit equations:
Accounting Profit = II – IE – OC, and Economic Profit = II – IE – OC – Q
where II represents interest expense, IE interest income, OC operating costs (which include impairment charges on bad debt), and Q liquidity cost.
For a bank to remain economically profitable (or even viable in the long-term), the rate of economic profit must be at least equal to the bank’s cost of equity.
To maximise their economic profits, banks look for the most-profitable risk-adjusted lending opportunities. ‘Lending’ to the ECB is one of those opportunities. Placing money at the ECB generates interest income. This interest income is more than welcome to (at least) maintain some level of accounting profitability (though not necessarily economic profitability) when economic conditions are bad and income from lending drops while impairment charges jump***.
With deposit rates at 0, banks’ income became fully constrained by financial markets and the economy. With rates in negative territory, not only banks see their interest income vanish but also their interest expense increase. From the equations above, it is clear that it makes banks less profitable****. On top of that, lending that cash can make banks less liquid, which increases their riskiness and elevates their cost of capital (the ‘Q’ above). The question becomes: adjusted for credit and liquidity risk, is it still worth keeping that cash at the ECB? The answer is probably yes.
(unless banks find worthwhile investments outside of the Eurozone, which wouldn’t be of much help to prop up Euro economies…)
To summarise, ‘II’ is negatively impacted by a low base rate whereas ‘IE’ reaches a floor (= margin compression). ‘IE’ then increases when the central bank deposit rate turns negative. Meanwhile, ‘OC’ increases as loan impairment charges jump due to heightened credit risk. Profitability is depressed, partly due to the central bank’s decisions.
Many European banks aren’t currently lending because they are trying to implement new regulatory requirements (which makes them less profitable) in the middle of an economic crisis (which… also makes them less profitable). As a result, the ECB measures seem counterproductive: in order to lend more, banks need to be economically profitable. Healthy banks lend, dying ones don’t.
The ECB is effectively increasing the pressure on banks’ bottom line, hardly a move that will provide a boost to lending. The only option for banks will be to cut costs even further. And when a bank cut costs, it effectively reduces its ability to expand as it has less staff to monitor lending opportunities, and consequently needs to deleverage. Once profitability is re-established, hiring and lending could start growing again.
A counterintuitive (and controversial) approach to provide a boost to lending would be to subsidise even more the banking sector by increasing interest rates on both the refinancing and deposit facilities.
Defining the appropriate level of interest rates would be subtle work though: struggling over-indebted households and businesses may well start defaulting on their debt. On the other hand banks’ revenues would increase as margin compression disappears, making them able to lend more eventually. The subtle balance would be achieved when interest income improvements more than offset credit losses increases. Not easy to achieve, but pushing rates ever lower is likely to cripple the banking system ever more and reduce lending in proportion (while allowing zombie firms to survive).
Furthermore, banks are repricing their loan book upward anyway, making the ECB rate cuts pointless. The process takes time though and it would be better for banks to rebuild their revenue stream sooner than later. The ECB could still use other monetary tools to influence a range of interest rates and prices through OMO and QE measures, which would be less disruptive to banks’ margins.
Finally, the ECB has launched its own-FLS style ‘TLTRO’, a scheme that provides cheap funding to banks if they channel the funds to businesses. Similarly to the BoE’s FLS, I believe such scheme suffers from delusion. Banks are currently deleveraging to lower their RWAs in order to comply with the harsher capital requirements of Basel 3. If there is one thing banks want to avoid, it is to lend to RWA-dense customers such as SMEs… (and instead focus on better RWA/risk-adjusted profitable lending such as… mortgages). Banks can also already extract relatively low wholesale funding rates by issuing secured funding instruments such as covered bonds. (UPDATE: see this follow-up post on that topic)
* This does not mean that banks would ‘lend out’ money to customers, unless they withdraw it as cash. But by increasing lending, absolute reserve requirements increase and banks have to transfer money from the deposit facility to the reserve facility.
** Data comes from respective central banks. They are not fully comparable. I gathered data from many different Eurozone countries, but unfortunately, some central banks don’t provide the data I need (or the statistics database doesn’t work, as in Italy…). Spreads are approximate ones calculated between the middle point of deposit rates and the middle point of lending rates. Analysis is very superficial, and for a more comprehensive methodology please refer to my equivalent posts on the UK/BoE.
*** Don’t get me wrong though. Fundamentally speaking, I am not in favour of such central bank mechanisms as I believe this is akin to a subsidy that distorts banks’ risk-taking behaviour. In a central banking environment, like Milton Friedman I’d rather see the central bank manipulate interest rates solely through OMO-type operations.
**** Of course this remains marginal. But in crisis times, ‘marginal’ can save a bank. Let’s also not forget booming litigation charges, currently estimated at USD104Bn… Evidently making it a lot easier for banks to lend as you can imagine…
Two brand new pieces of academic research have been published last month, directly or indirectly related to the Austrian theory of the business cycle (some readers might already know my RWA-based ABCT: here, here, here and here).
The first one, called Roundaboutness is Not a Mysterious Concept: A Financial Application to Capital Theory (Cachanosky and Lewin) attempts to start merging ABCT (or rather, Austrian capital theory) with corporate finance theory. The authors use the finance concepts of economic value added (EVA), modified duration, Macaulay duration and convexity in order to represent the Austrian concepts of ‘roundaboutness’ and ‘average period of production’. The paper provides a welcome and well-defined corporate finance background to the ABCT.
However, finance practitioners still don’t have the option to use a ‘full-Austrian’ alternative financial framework, as this paper still relies on some mainstream concepts. For instance, the EVA calculation for a given period t is as follows:
where ROIC is the return on invested capital, WACC the weighted average cost of capital and K the financial capital invested.
In order to compute the project’s market value added (MVA, i.e. whether or not the project has added value), it is then necessary to discount the expected future EVAs of each period t1, t2…, T, by the WACC of the project:
The WACC represents the minimum return demanded by investors to compensate for the risk of such a project (i.e. the opportunity cost), and is dependent on the interest rate level. The problem arises in the way it is calculated in modern mainstream finance. While the cost of debt capital is relatively straightforward to extract, the cost of equity capital is commonly computed using the capital asset pricing model (CAPM). Unfortunately, the CAPM is based on the Modern Portfolio Theory, itself based on new-classical economics and rational expectations/efficient market hypothesis premises, which are at odds with Austrian approaches.
(And I am not even mentioning some of the very dubious assumptions of the theory, such as “all investors can lend and borrow unlimited amounts at the risk-free rate of interest”…)
While it is easy for researchers to define a cost of equity for a theoretical paper, practitioners do need a method to estimate it from real life data. This is how the CAPM comes in handy, whereas the Austrian approach still has no real alternative to suggest (as far as I know).
Nevertheless, putting the cost of equity problem aside, the authors view the MVA as perfectly adapted to capital theory:
Note that the MVA representation captures the desired characteristics of capital-theory; (1) it is forward looking, (2) it focuses on the length of the EVA cash-flow, and (3) it captures the notion of capital-intensity.
Using the corporate finance framework outlined above, the authors easily show that the more capital intensive investments are the more they are sensitive to variations in interest rates (i.e. they have a larger ‘convexity’). They also show that more ‘roundabout’/longer projects benefit proportionally more from a decline in interest rates than shorter projects. Unsurprisingly, those projects are also the first ones to suffer when interest rates start going up.
The following chart demonstrates the trajectory of the MVA of both long time horizon (high roundabout – HR) and short time horizon (low roundabout – LR) projects as a function of WACC.
Overall, this is a very interesting paper that contains a lot more than what I just described. I wish more research was undertaken on that topic though.
The second paper, pointed by Tyler Cowen, while not directly related to the ABCT, nonetheless has several links to it (I am unsure why Cowen thinks this piece of research actually reflects the ABCT). What’s interesting in this paper is that it seems to confirm the link between credit expansion, financial instability and banks stock prices, as well as the ‘irrationality’ of bank shareholders, who do not demand a higher equity premium when credit expansion occurs (which doesn’t seem to fit the rational expectations framework very well…).
The ‘cut the middle man’ effect of P2P lending is already celebrated for offering better rates to both lenders and borrowers. But what many people miss is that this effect could also ease the transmission mechanism of central banks’ monetary policy.
I recently explained that the banking channel of monetary policy was limited in its effects by banks’ fixed operational costs. I came up with the following simplified net profit equation for a bank that only relies on interest income on floating rate lending as a source of revenues:
Net Profit = f1(central bank rate) – f2(central bank rate) – Costs, with
f1(central bank rate) = interest income from lending
= central bank rate + margin and,
f2(central bank rate) = interest expense on deposits
= central bank rate – margin
(I strongly advise you to take a look at the details here, which was a follow-up to my response to Ben Southwood’s own response on the Adam Smith Institute blog to my original post…which was also a response to his own original post…)
Consequently, banks can only remain profitable (from an accounting point of view) if the differential between interest income and interest expense (i.e. the net interest income) is greater than their operational costs:
Net interest income >= Costs
When the central bank base rate falls below a certain threshold, f2 reaches zero and cannot fall any lower, while f1 continues to decrease. This is the margin compression effect.
Above the threshold, the central bank base rate doesn’t matter much. Below, banks have to increase the margin on variable rate lending in order to cover their costs. This was evidenced by the following charts:
As the UK experience seems to show, banks stopped passing BoE rate cuts on to customers around a 2% BoE rate threshold. I called this phenomenon the ‘2%-lower bound’. I have yet to take a look at other countries.
Enter P2P lending.
By directly matching savers and borrowers and/or slicing and repackaging parts of loans, P2P platforms cut much of banks’ vital cost base. P2P platforms’ online infrastructure is much less cost-intensive than banks’ burdensome branch networks. As a result, it is well-known that both P2P savers and borrowers get better rates than at banks, by ‘cutting the middle man’. This is easy to explain using the equations described above, as costs approach zero in the P2P model. This is what Simon Cunningham called “the efficiency of Peer to Peer Lending”. As Simon describes:
Looking purely at the numbers, Lending Club does business around 270% more efficiently than the comparable branch of a major American bank
Simon calculated the ‘efficiency’ of each type of lender by dividing the outstanding loans of Wells Fargo and Lending Club by their respective operational expenses (see chart below). I believe Lending Club’s efficiency is still way understated, though this would only become apparent as the platform grows. The marginal increase in lending made through P2P platforms necessitates almost no marginal increase in costs.
Perhaps P2P platforms’ disintermediation model could lubricate the banking channel of monetary policy the closer central banks’ base rate gets to the zero bound?
Possibly. From the charts above, we notice that the spread between savings rates and lending rates that banks require in order to cover their costs range from 2 to 3.5%. This is the cost of intermediation and maturity transformation. Banks hire experts to monitor borrowers and lending opportunities in-house and operate costly infrastructures as some of their liabilities (i.e. demand deposits) are part of the money supply and used by the payment system.
However, disintermediated demand and supply for loanable funds are (almost) unhampered by costs. As a result, the differential between borrowers and savers’ rate can theoretically be minimal, close to zero. That is, when the central bank lowers its target rate to 0%, banks’ deposit rates and short-term government debt yield should quickly follow. Time deposits and longer-dated government debt will remain slightly above that level. Savers would be incentivised to invest in P2P if the proposed rate at least matches them, adjusting for credit risk.
Let’s take an example: from the business lending chart above, we notice that business time deposit rates are currently quoted at around 1%. However, business lending is currently quoted at an average rate of about 3%. Banks generate income from this spread to pay salaries and other fixed costs, and to cover possible loan losses. Let’s now imagine that companies deposit their money in a time deposit-equivalent P2P product, yielding 1.5%. Theoretically, business lending could be cut to only slightly above 1.5%. This represents a much cheaper borrowing rate for borrowers.
P2P platforms would thus more closely follow the market process: the law of supply and demand. If most investments start yielding nothing, P2P would start attracting more investors through arbitrage, increasing the supply of loanable funds, and in turn lowering rates to the extent that they only cover credit risk.
The only limitation to this process stems from the nature of products offered by platforms. Floating rate products tend to be the most flexible and quickly follow changes in central banks’ rates. Fixed rate products, on the other hand, take some time to reprice, introducing a time lag in the implementation of monetary policy. I believe that most P2P products originated so far were fixed rate, though I could not seem to find any source to confirm that.
In the end, P2P lending is similar to market-based financing. The bond market already ‘cuts the middle man’, though there remains fees to underwriting banks, and only large firms can hope to issue bonds on the financial markets. In bond markets, investors exactly earn the coupon paid by borrowers. There is no differential as there is no middle man, unlike in banking. P2P platforms are, in a way, mini fixed-income markets that are accessible to a much broader range of borrowers and investors.
However, I view both bond markets and P2P lending as some version of 100%-reserve banking. While they could provide an increasingly large share of the credit supply, banks still have a role to play: their maturity transformation mechanism provides customers with a means of storing their money and accessing it whenever necessary. Would P2P platform start offering demand deposit accounts, their cost base would rise closer to that of banks, potentially raising the margin between savers and borrowers as described above.
It seems that, by partly shifting from the banking channel to the P2P channel over time, monetary policy could become more effective. I am sure that Yellen, Carney and Draghi will appreciate.
Following my recent reply to Ben Southwood on the relationship between mortgage rates, BoE base rate and banks’ margins and profitability (see here and here), a question came to my mind: if the BoE rate can fall to the zero lower bound but lending rates don’t, should we still speak of a ‘zero lower bound’? It looks to me that, strictly in terms of lending and deposit rates, setting the base rate at 0% or at 2% would have changed almost nothing at all, at least in the UK.
The culprit? Banks’ operational expenses. Indeed, it looks like the only way to break through the ‘2%-lower bound’ would be for banks to slash their costs…
Let’s take a look at the following mortgage rates chart from one of my previous posts:
From this chart, it is clear that lowering the BoE rate below around 2.5% had no further effect on lowering mortgage rates. As described in my other posts, this is because banks’ net interest income necessarily has to be higher than expenses for them to remain profitable. When the BoE rate falls below a certain threshold that represents operational expenses, banks have to widen the margins on loans as a result.
What about business lending rates? Since business lending is funded by both retail and corporate deposits (and excluding wholesale funding for the purpose of the exercise), the analysis must take a different approach. Banks don’t often disclose the share of corporate deposits within their funding base, but I managed to find a retail/corporate deposit split of 75%/25% at a large European peer, which I am going to use as a rough approximation to estimate banks’ business lending margins. Here are the results of my calculations (first chart: margin over time deposits, second chart: margin over demand deposits):
No surprise here, the same margin compression effect appears as a result of the BoE rate collapsing (as well as Libor, as floating corporate lending is often calculated on a Libor + margin basis, unlike mortgages, which are on a BoE + margin basis). Before that period, changes in the BoE and Libor rates had pretty much no effect on margins. After the fall, banks tried to rebuild their margins by progressively repricing their business loan books upward (i.e. increasing the margins over Libor).
Here again we can identify a 1.5% BoE rate floor, under which lowering the base rate does not translate into cheaper borrowing for businesses:
This has repercussions on monetary policy. The banking/credit channel of monetary policy aims at: 1. easing the debt burden on indebted household and businesses and, 2. stimulating investments and consumption by making it cheaper to borrow. However, it seems like this channel is restricted in its effectiveness by banks’ ability in passing the lower rate on to customers. Banks’ short-term fixed cost base effectively raises the so-called zero lower bound to around 2%. The only way to make the transmission mechanism more efficient would be for banks to drastically improve their cost efficiency and have assets of good-enough quality not to generate impairment charges, which is tough in crisis times. Unfortunately, there are limits to this process, and a bank without employee and infrastructure is unlikely to lend in the first place…
Don’t get me wrong though, I am not saying that lowering the BoE rate (and unconventional monetary policies such as QE) is totally ineffective. Lowering rates also positively impact asset prices and market yields, ceteris paribus. This channel could well be more effective than the banking one but it isn’t the purpose of this post to discuss that topic. Nevertheless, from a pure banking channel perspective, one could question whether or not it is worth penalising savers in order to help borrowers that cannot feel the loosening.
PS: I am not aware of any academic paper describing this issue, so if you do, please send me the link!