One of the little known facts of the history of monetary policy is that until 1994, the Fed did not actually announce interest rate decisions. Market participants had to infer rate changes from the Fed’s open market operations. This is just one example of how so many things we take to be natural and obvious are, in reality, relatively recent phenomena. In less than twenty years, the Fed has transitioned from near-opacity to an almost obsessive transparency. Another example of a relatively recent monetary policy doctrine that is now unquestioned is the doctrine of inflation targeting. The essential idea of inflation targeting is that people and firms should not have to think about the level and volatility of inflation when they make economic decisions. Inflation must therefore be kept at low and stable levels so that the long-run costs of unpredictable and uncertain inflation are minimised. As Mervyn King notes, inflation targeting has always been about improving the “credibility and predictability of monetary policy”.
However, in a world where money earns interest, minimising the uncertainty of macroeconomic policy does not equate to minimising the volatility of inflation. When all money bears interest, all that matters for those who hold money or bonds is the real interest rate earned on money and bonds. Given the fiscal stance and state of private credit growth, central banks should manage the real rate of interest such that rentiers do not capture a free lunch (i.e. real rates should not be too high) and there is no risk of a hot-potato/credit-bubble cycle (i.e. real rates should not be too low).
Money does not bear interest today because central banks pay interest on reserves. The primary reason why we live in a world of interest-bearing money is the gradual deregulation and innovation in financial markets over the last thirty years that triggered a shift from money to near-money assets. Apart from minimal liquidity reserves, there is simply no need to hold significant amounts of money in one’s zero-interest current account. Individuals can hold money in money market funds or treasury ETFs. Firms and high net-worth individuals can simply hold treasury bills that are as risk-free and liquid as money is. Even treasury bonds consist of a risk-free component that can be separated from the duration-risk component and monetised via the repo market. The equivalence of money and bonds is not just a temporary “liquidity trap” phenomenon. The evolution of financial markets means that the role of interest-free money is obsolete, now and forever.
In such an environment, the uncertainty and the volatility that individuals and firms care about is the volatility of the real interest rate. Let me take a simple example to illustrate this point. Let us assume that you hold a significant proportion of your assets in a short-term T-bill ETF that currently yields 0% in an environment when inflation is 2%. Therefore, the real interest rate is -2% but you swallow this loss as a “safety premium” fearful that investing in risky assets inflated by monetary stimulus may result in much greater losses. Now suppose the Fed decides to adopt an inflation target of 5% instead, which it achieves by buying up private sector assets such as equities1 while still holding the Fed Funds rate at 0%. This move to 5% inflation obviously hurts your investment in T-bills but the real reason is not that inflation has gone up. The real reason is that real rates have turned even more negative from -2% to -5%.
Many economists will complain that there is no other option. But when government bonds effectively function as money, there are a multitude of other options. The 5% inflation target could be hit by instituting a significant increase in fiscal stimulus (preferably via helicopter drops) and simultaneously hiking rates to 3%. From the perspective of most firms and individuals, this option which minimises the volatility of real interest rates is by far the more predictable and less uncertain policy. Those who borrow or invest at fixed rates for longer tenors will obviously suffer more volatility but such activities are explicitly risk-taking by nature. There is no conceivable reason why the central bank or the government should subsidise such risk-taking.
Proponents of inflation-targeting sometimes point to the poor economic performance of many developing countries with high and variable inflation. But what is really at fault in many of these instances is the tendency of the fiscal and monetary authorities to inject unexpected bursts of inflation that are uncompensated for by the interest rate regime enforced by the central bank. It is the persistent erosion of purchasing power due to negative real interest rates2 that is the real source of the poor macroeconomic performance in most high-inflation regimes.
The obvious object to my argument is that there is no reason why real interest rates must be held constant – I agree. My argument is not that real rates should be held constant but simply that excessive volatility in real rates must be avoided even if it is at the expense of a more volatile inflation rate. If the economy is hit by an inflationary supply shock, then it must be met by an increase in the inflation rate and an increase in the nominal interest rate (thus keeping real rates stable) rather than a rate hike to maintain a constant inflation rate (which would simply be an unwarranted transfer of wealth to lenders). There are also limits to how negative real rates can be driven to before an inflationary spiral is triggered. As of now, we are clearly well within these limits but it is foolish to assume that the inflation target can be increased to any level while central bank rates still remain at zero. Sooner or later, increasingly negative real rates will set off a inflationary spiral and a stampede to buy up real assets instead of nominal bonds.
I am not opposed to central banks driving down real rates to counter the increased demand for safety during a liquidity crisis. But in an environment like today when junk bond prices are at all-time highs, there is no justification for maintaining artificially low real rates. To restore economic prosperity in today’s crony capitalist and stagnant economy, we need to provide direct transfers to individuals via money-financed stimulus while simultaneously hiking rates to stop the permanent bailout of incumbent banks and firms.
See my earlier post ‘The Case Against Monetary Stimulus Via Asset Purchases’ for why I oppose such a policy. ↩
Negative rates may also be enforced by restrictions on the interest rate payable on deposits and similar financial market regulations. ↩
I have written a new essay in the economics section of my new project reconciling the possibility of technological unemployment during a time of innovative stagnation titled Technological Unemployment Amidst Stagnation. Most of it is just a rehash of arguments that I have made on this blog especially in the posts Innovation, Stagnation and Unemployment and Advances in Technology and Artificial Intelligence: Implications for Education and Employment but about 25-30% of the essay is new material and the overall essay itself is written in a much simpler manner than many of my posts here.
Do let me know what you think. Thanks for reading.
Most of my posts on this blog tackle different topics from essentially the same perspective, which is a biological/ecological systems-influenced take on the concept of system resilience. But the basic concepts of this underlying framework are spread out among a bunch of overlapping posts. In an attempt to make each post stand on its own, I find myself spending more and more time repeating past arguments in new posts in order to set the stage for new ideas. I find this repetition incredibly boring and at the same time, this half-hearted repetition probably doesn’t do much to help a new reader either.
To try and resolve this problem, I’m going to write most of my interdisciplinary resilience material in a more hierarchical format. The format I’m going to adopt here is to distil the essential framework into short summary essay and organise the rest of my writing by domain – economics, biology, ecology, technology etc. The idea is that you should be able to read the summary essays and then dive into whichever domain takes your fancy. Conversely the core ideas explored in each domain should mostly be summarised in the summary essays. These essays will be relatively simple but the material within the domain sections will be much more detailed and often messy.
This format also allows me to reconcile my desire to draw generalised parallels across domains while at the same time exploring each domain in detail. So to kick off this experiment, I’ve posted the initial summary essay titled “All Systems Need A Little Disorder” at http://alittledisorder.com/all-systems-need-a-little-disorder/. The basic idea in this essay – that systems that need to be robust and generate novelty need some chaos (not a lot of chaos, just a little bit) – underlies pretty much all of the posts I have written on this blog. In the next few months, I will apply and elaborate upon the basic concepts in this essay across a wider array of domains than I have explored so far.
By adopting this format, I can hopefully increase the output of material that I post online and explore new topics more easily without having to rehash old posts. At the same time new readers will hopefully have a more coherent way of navigating through my ideas. I will post links to any new material on this blog as well for those of you who subscribe to the RSS feed here. I will continue to write here but the material will probably be more limited to macroeconomics and finance than it has previously been.
As always any feedback is much appreciated. Thanks for reading.
In an earlier post, I argued that our current monetary system is close to being to a Wicksellian ‘pure credit economy’. In Hans Trautwein’s words, this is “a state of affairs in which all money is held in interest- bearing bank deposits and in which all payments are effected by means of book-keeping transfers in the banking system”. One significant way in which our current system is not quite a pure credit economy is that economic agents still retain the option to hold currency notes. This option is not very important in positive-rate environments but it denies the central bank the ability to enforce negative interest rates (which can be avoided by simply hoarding zero-interest physical notes). The dominance of interest-bearing money combined with the inability to enforce negative interest rates implies that the quantity of base money in the system is irrelevant, not just now in a ‘liquidity trap’, but at all points in the future.
The Irrelevance of The Quantity of Base Money and The Absence of The Monetarist Hot Potato
It is trivially obvious that interest-bearing money cannot be a hot potato in the monetarist sense. There is no reason to get rid of interest-bearing money balances and interest-bearing money holdings only need to be minimised if the interest rate is insufficient relative to the ‘natural’ real rates and safety premium implied in holding money. To put it simply, if interest rates are 5% and inflation is 15% then interest-bearing money will act as a hot potato and fuel inflation. But in the current environment of possibly negative natural real rates and a high demand for safety, prolonged negative real rate regimes are perfectly sustainable without triggering any ‘hot potato’ inflation. The above holds not only at the zero-bound but at all positive interest rates. If the central bank wants to sustain positive bank rates, it must either pay interest on reserves or mop up all excess reserves. In either scenario, we have no hot potato.
Interest-Bearing Money: Debt as Money
The history of interest-bearing money is essentially the history of debt as money. The modern history of transferable debt as money is exemplified by the use of bills of exchange in post-Renaissance Europe. As Philip Coggan explains:
trading systems were an early form of our modern economy, with its layers of debt and reliance on paper money. A merchant might extend credit to his customers; in turn, he would need such credit from his own suppliers, who might only have bought the goods with money borrowed from someone else. The default of one party would ripple through the system. This system was formalized in the form of bills of exchange, promissory notes offered as payment from one trader to another. The recipient might then use the bill as collateral to raise cash from a bank or other lender. The bill would be accepted at a discount, depending on a number of factors, most crucially the creditworthiness of the merchant concerned. This was, in effect, a paper money system outside the government’s control.
It is instructive to examine the evolution of this private credit economy in order to fully understand where we stand now. The rest of this section is primarily drawn from Carl Wennerlind’s excellent book ‘Casualties of Credit’. The private credit economy was an essential component of the English economy due to the perennial shortage of metallic currency. As Wennerlind notes, it wasn’t just merchants but the bulk of the English population who were participants in the credit economy. In the early days in the seventeenth century , the supply of private credit alone was nowhere near enough to make up for the scarcity of metallic currency. One reason was the limited transferability of private debt, a problem that was solved by the passage of the Promissory Notes Act of 1704 which made all debt instruments negotiable. Nevertheless, the limited elasticity of the private credit system in responding to demand from commerce remained a problem. A related and equally severe problem was fragility induced by the possibility of default. This was the Achilles heel of the private credit economy in the 17th century and it remains the case in the 21st century. Just as in the 21st century, the real systemic risk is the threat of a wave of cascading defaults brought upon by the tightly interconnected nature of private credit agreements.
The solution to this problem that we are all aware of and that has been well-documented is the growth of modern banking ultimately backstopped by central banks (usually via lender-of-last-resort actions). This architecture was initially limited by the restrictions placed upon the central bank by the metallic/gold standard, Bretton Woods etc which were finally thrown away in 1971 to construct the “perfectly elastic” monetary system that we have today. This is the logical conclusion of the process of abstracting away from the prior personal nature of “money as debt” to a decentralised impersonal system.
Less documented but equally important are the attempts to improve the supply and safety of the credit economy via collateral. The idea is simple – assets can be used as security to back the credit, thus improving the supply as well as the safety of credit. There were many recommendations as to what constitutes eligible collateral in the 17th and 18th century but by far the most popular suggestion was land. It is worth quoting Wennerlind on this subject (who in turn quotes William Potter):
Potter also offered a proposal for a land bank, which was remarkably similar to that of Culpeper. Since “Credit grounded upon the best security is the same thing with Money,” the key was to establish a bank that used a different asset than precious metals as security backing the credit money. Since land was considered the most concrete and stable commodity at the time, there could be no better security than land to induce people to part with their commodities in exchange. By mortgaging land, which “would serve as well and better for such a pawn,” the land bank created a credit currency that would have “as true intrinsick value, as Gold and Silver”
Others, such as Hugh Chamberlen advocated a general storehouse of goods that would serve as collateral. Nevertheless, none of these ideas were adopted in 17th/18th century England for good reasons – none of these choices for collateral were liquid enough or permanent enough for the purpose. To a modern investor, government bonds are the obvious answer to this dilemma. But in 17th century Europe, government debt was neither liquid nor safe. However, a series of institutional changes after the ‘Glorious Revolution’ in 1688 changed all this (see North and Weingast 1989 for details). With the setting up of the Bank of England, British government bonds began to resemble the “risk-free” counterparts of the modern world by the mid-18th century. It is obvious how the ability of the new more representative English Parliament to credibly commit to repay its debts enabled England to fund itself at a much lower cost. What is less appreciated is the fillip that the institution of a liquid, comparatively safe government bond market gave to the private credit economy. As Baskin and Miranti note, these government obligations could be used to collateralise private borrowing in a manner that is uncannily similar to the modern-day term repo contract.
The Hot Potato Constraint in a Credit Economy
What does all this have to do with the modern monetary system? In the modern pure fiat-currency economy (i.e. not the Eurozone), interest-bearing deposits, interest-bearing central bank reserves and interest-bearing government debt are all equivalent in that they are all nominally safe state obligations unencumbered by restraints such as a gold standard. Any shift in liabilities between central bank reserves, deposits and debt engineered by the central bank is only relevant for its interest-rate impact. There is nothing in this process that can be even remotely termed as “money printing”. The inflation tax and any “hot potato” effect are dependent not on the absolute levels of inflation but the real interest rate offered on each tenor of these government obligations.
To the extent that any activity of the state approaches money printing, it is the act of deficit spending. Even this does not necessarily entail inflation – the central bank can force a contraction in the private credit economy by a sufficient rate-hike to counter any fiscal stance. Again there is no inflation tax and no possibility of hyperinflation as long as interest rates across the government obligation curve compensate sufficiently for inflation. Each fiscal stance has a separate sustainable level of inflation and interest rates that constitutes a short-term equilibrium. When a loose fiscal stance breaks out into excessive inflation and the risk of hyperinflation, it is usually the result of this rate hike being inadequate for fear of a collapse in the private economy.
Rather than talk in the abstract, it is easier to elaborate on the above framework with a few relevant and timely examples.
Permanence of QE is irrelevant
Gavyn Davies gives us the conventional argument as to why the perceived temporary nature of QE matters in preventing out-of-control inflation:
Fiscal policy, in theory at least, is set separately by the government, and the budget deficit is covered by selling bonds. The central bank then comes along and buys some of these bonds, in order to reduce long-term interest rates. It views this, purely and simply, as an unconventional arm of monetary policy. The bonds are explicitly intended to be parked only temporarily at the central bank, and they will be sold back into the private sector when monetary policy needs to be tightened. Therefore, in the long term, the amount of government debt held by the public is not reduced by QE, and all of the restraining effects of the bond sales in the long run will still occur. The government’s long-run fiscal arithmetic is not impacted.
As I have illustrated above, QE in a world of interest-bearing money is simply an adjustment in the maturity profile of government debt. But that is not all. In a credit economy where government bonds are a repoable safe asset, the bond-holder can simply repo his bonds for cash if he so chooses. Just as the East India Company could access cash on the back of their government bond holdings in the 18th century, any pension fund, insurer or bank can do the same today. This illustrates why the reversal of QE, if and when it happens, will have no impact on economy-wide access to cash/purchasing power.
Bond-financed or Money-financed deficits
Gavyn Davies again gives us the conventional argument:
When it runs a budget deficit, the government injects demand into the economy. By selling bonds to cover the deficit, it absorbs private savings, leaving less to be used to finance private investment. Another way of looking at this is that it raises interest rates by selling the bonds. Furthermore the private sector recognises that the bonds will one day need to be redeemed, so the expected burden of taxation in the future rises. This reduces private expenditure today. Let us call this combination of factors the “restraining effect” of bond sales.
All of this is changed if the government does not sell bonds to finance the budget deficit, but asks the central bank to print money instead. In that case, there is no absorption of private savings, no tendency for interest rates to rise, and no expected burden of future taxation. The restraining effect does not apply. Obviously, for any given budget deficit, this is likely to be much more expansionary (and potentially inflationary) than bond finance.
The ability of the private sector to repo its government bonds to access purchasing power today gives us a profound result. Whether the central bank monetises government debt or not is almost irrelevant (except from a signalling perspective) because the private sector can monetise government debt just as effectively. And when the government debt does not represent a ‘hot potato’, the private sector often does exactly that. This is not a theoretical argument. For example, Akçay et al illustrate how fiscal deficits led to inflation in Turkey despite the absence of monetisation because ” innovations in the form of new financial instruments are encouraged through high interest rates, and repos are typical examples of such innovations in chronic and high inflation countries. People are thus able to hold interest-bearing assets that are almost as liquid as money, and monetization is effectively done by the private financial sector instead of the government”. As Çavuşoğlu summarises, “The money creation process under high budget deficits can as well be characterised as an endogenous credit-money expansion rather than a monetary expansion to maximize seignorage revenue”.
Lest you assume that this only applies to developing market economies, the same argument has been made almost three decades ago by Preston Miller:
In the financial sector….higher interest rates make profitable the development of new financial instruments that make government bonds more like money. These instruments allow people to hold interest-bearing assets that are as risk-free and as useful in transactions as money is. In this way, the private sector effectively monetizes government debt that the Federal Reserve doesn’t, so the inflationary effects of higher deficit policies increase.
Even in the early 80s, Miller saw the gradual demise of non-interest bearing money:
In recent years in the United States there have developed, at money market mutual funds, demand deposit accounts that are backed by Treasury securities and, at banks, deep-discount insured certificates of deposit that are backed by Treasury securities, issued in denominations of as little as $250, and assured of purchase by a broker. In Brazil, which has run high deficits for years, Treasury bills have become very liquid: their average turnover is now less than two days.
As in the case of Turkey and as argued by Preston Miller, the private sector can monetize the deficit as effectively as the central bank can. And so long as government obligations are deemed safe, it almost certainly will. In an interest-bearing economy, the safety of these obligations have nothing to do with the absolute level of inflation and everything to do with the real rate of return on the bonds. When central banks and governments attempt to enforce an excessively negative rate of return, they play with fire and risk hyperinflation.
The Near-Permanence of (Non Hot-Potato) Government Debt
P.G.M. Dickson characterised the rise of the government bond market in London during the 18th century as the era of “debts that were permanent for the state, liquid for the individual”. In a credit economy, government debt issued in the past is simply money that has already been printed. Erasing this debt would not imply a monetary collapse but it would unleash strong deflationary forces.
Most of the developed world (ex the Eurozone) could easily maintain their current levels of government debt ad infinitum so long as the real interest rates paid on them are sufficient. And in fact it makes sense for them to do exactly that. Even without the monetisability of long-term government debt, there is a significant demand for them from many private sector holders – the pension fund and insurance industry which needs long-tenor bonds to match its liabilities to retirees, investors who need long-tenor bonds to hedge their risky assets and provide tail-risk protection. Even without taking into account the “natural” real rate of interest, there is a strong argument to be made that the average real rate of return on long-tenor government bonds should be negative. Therefore, it does not even make economic sense for governments to pay back their debt, as long as it can be serviced at a sustainable real rate.
The appropriate question to ask is not ‘What is the maximum level of government debt is that can be plausibly paid back?’. It is ‘What is the maximum level of government debt that can be plausibly serviced on a permanent basis?’. If there are any Ponzi schemes in government debt, they exist only if and when there are real limits to economic growth – working-age population growth, energy limits etc.
What Matters: Future Deficits and Real Rates
A policy option such as a cancellation of past debt or an announcement of helicopter drops would be relevant to the extent that it effects future deficits. Higher deficits would typically warrant a more hawkish monetary stance and it is the combination of this fiscal stance and the monetary response that determines whether the deficit regime constitutes an inflation tax on the private sector. For example, the state could institute a helicopter drop and raise interest rates at the same time to maintain real rates at acceptable levels – again the level of real rates is much more important than the absolute level of inflation. Even this hike in rates may not be required if the private sector is undergoing an endogenous delevering at the time.
Modern Repo and the Asset Price Approach to Monetary Policy
If the collateral underpinning the private credit economy was limited to government bonds, the lender-of-last-resort role of the central bank in the repo market would be trivial. However, the current scope of the repo market and similar financing arrangements (notably ABCP) extends to far riskier assets. Although the risk management in today’s repo market is far superior from an individual counterparty’s perspective ( the predominance of the overnight repo, more sophisticated margining etc ) the systemic risk of cascading defaults triggering a credit collapse has in fact spread to all asset markets.
In order to meet their stabilisation mandate, central banks have implicitly taken on a mandate to backstop and stabilise the entire spectrum of liquid asset markets. If the central banks influence anything that could be termed as money supply in the modern credit economy, they do so via their influence on asset price levels (influenced in turn through the central bank’s actions on present interest rates, future interest rate path and liquidity). In a collateral-dependent credit economy, the Greenspan Put is the logical end-point of the stabilisation processes, the modern motto of which could be summarised as: ‘Focus on collateral values and the money supply will take care of itself’. Successive stabilisation leaves the economy in a condition where all economic actors have moved away from the idiosyncratic, illiquid economic risks that are the essence of an innovative, entrepreneurial economy towards the homogeneous liquid risks of a stagnant economy (detailed argument here).
In an earlier post, I noted that “The long-arc of stabilised cycles is itself a disequilibrium process (a sort of disequilibrium super-cycle) where performance in each cycle deteriorates compared to the last one – an increasing amount of stabilisation needs to be applied in each short-run cycle to achieve poorer results compared to the previous cycle.” This sentiment applies even when we look at the long-arc of stabilisation in England since the 17th century. In the 17th century, it only took a change in the laws (making debts negotiable) to prevent a collapse in the credit economy whereas now we need to prop up the entire spectrum of asset markets.
1. The section on monetary hot potatoes and high-powered money is almost completely taken from commenter ‘K’ – example here
Hyperinflation is often viewed as a phenomenon where a currency is repudiated by its holders who refuse to hold the currency in any nominal form i.e. a collapse in demand for the currency. This is a reasonable but nevertheless incomplete explanation of how hyperinflation plays out in reality in modern capitalist economies.
In a world of interest-bearing money, high and monetised fiscal deficits by themselves are not reason enough for deposit-holders to repudiate a currency. If the central bank maintains a nominal interest rate at the short-end that compensates money-holders for the fiscally-created inflation, then there is no shortage of willing holders for the currency (in an interest-bearing form such as deposits). By the same token, the central bank must adjust lending rates at the short end (the equivalent of the ECB repo rate) to avoid an explosion in private credit growth fuelled by negative real rates. For each chosen fiscal regime, there is a monetary stance that can avert hyperinflation. However as fiscal deficits increase beyond a point, the equilibrating monetary stance consists of a nominal lending rate that must necessarily crowd out the private economy.
Understandably, most central banks are reluctant to raise rates in such a dramatic fashion. Instead they raise rates but only to the extent that real rates remain negative but not negative enough to motivate a wholesale repudiation of the currency. So long as real rates are maintained at a small negative rate, deposit-holders usually treat it as a “safe asset” premium that they are willing to pay. An environment that enables hyperinflation as a possibility is triggered when these real rates turn significantly negative. Significant negative real rates encourage holders of the currency even in its interest-bearing deposit form to shed their holdings of the currency when faced with constant and high real losses, losses that cannot be justified simply on account of the safety premium of the nominal asset.
But this repudiation of money is not the core driver of hyperinflation as we know it in the modern world. What almost always accompanies this repudiation is a sustained barrage of borrowing at the artificially low nominal and real rate enforced by the central bank or government (directly or indirectly via banks). Until late in the Weimar inflation, the Reichsbank kept discount rates as low as 5% (see table here) , a free lunch that was taken full advantage of by bankers and industrialists to lever up and invest in any real assets they could find. As Adam Fergusson notes, “new borrowings from the Reichsbank…from whom commercial enterprises could obtain credit at very low discount rates even at the height of the crisis in 1923, were automatically written off” due to the ludicrously negative levels of real interest rates that were enabled by the Reichsbank. The same was true in Zimbabwe where the central bank not only maintained one-year treasury bill rates at a level well below the inflation rate (enabling monetisation of deficits at a subsidised rate) but did the same with prime and bank lending rates which led to the predictable explosion in private sector credit expansion (see data here for a sample month).
Once real rates become sufficiently negative, credit growth explodes and the positive feedback loop of ever higher inflation fuelled not just by currency repudiation but by active exploitation of the banking and central bank discount window to access essentially free loans is set in motion. In other words, hyperinflation in modern capitalist economies is characterised not just by a collapse in the demand for deposits but an explosion in demand for loans at the “free lunch” level of nominal interest rates enforced by the central bank. Many commentators have recently asserted that Iran is in the midst of hyperinflation. Whether this is actually the case is still unclear – the current bout of higher inflation and prices may yet turn out to be temporary. But what makes hyperinflation possible is clear. Both lending rates and deposit rates have been set at levels well below the inflation rate for years now, a situation that threatens to descend into farce with inflation at above 50% per month and bank rates at only 21% per annum.
In a perceptive post, Reihan Salam makes the point that private equity firms are simply an industrialised version of corporate America’s efficiency-seeking impulse. I’ve made a similar point in a previous post that the the excesses of private equity mirror the excesses of the economy during the neoliberal era. To right-wing commentators, neoliberalism signifies a much-needed transition towards a free-market economy. Left-wing commentators on the other hand lament the resultant supremacy of capital over labour and rising inequality. But as I have argued several times, the reality of the neoliberal transition is one where a combination of protected asset markets via the Greenspan Put, an ever-growing ‘License Raj’, regulations that exist primarily to protect incumbent corporates and persistent bailouts of banks and large corporates have given us a system best described as “stability for the classes and instability for the masses”.
The solution preferred by the left is to somehow recreate the golden age of the 50s and the 60s i.e. stability for all. Although this would be an environment of permanent innovative stagnation bereft of Schumpeterian creative destruction, you could argue that restoring social justice, reducing inequality and shoring up the bargaining position of the working class is more important than technological progress. In this post I will argue that this stability-seeking impetus is counterproductive and futile. A stable system where labour and capital are both protected from the dangers of failure inevitably breeds a fragile and disadvantaged working class.
The technology industry provides a great example of how disruptive competitive dynamics can give workers a relatively strong bargaining position. As Reihan notes, the workers fired by Steve Jobs in 1997 probably found employment elsewhere without much difficulty. Some of them probably started their own technology ventures. The relative bargaining power of the technology worker is boosted not just by the presence of a large number of new firms looking to hire but also by the option to simply start their own small venture instead of being employed. This vibrant ecosystem of competing opportunities and alternatives is a direct consequence of the disruptive churn that has characterised the sector over the last few decades. This “disorder” means that most individual firms and jobs are vulnerable at all times to elimination. Yet jobseekers as a whole are in a relatively strong position. Micro-fragility leads to macro-resilience.
In many sectors, there are legitimate economies of scale that prevent laid-off workers from self-organising into smaller firms. But in much of the economy, the digital and the physical, these economies of scale are rapidly diminishing. Yet these options are denied to large sections of the economy due to entry barriers from licensing requirements and regulatory hurdles that systematically disadvantage small, new firms. In some states, it is easier to form a technology start-up than it is to start a hair-braiding business. In fact, the increasingly stifling patent regime is driving Silicon Valley down the same dysfunctional path that the rest of the economy is on.
The idea that we can protect incumbent firms such as banks from failure and still preserve a vibrant environment for new entrants and competitors is folly. Just like a fire that burns down tall trees provides the opportunity for smaller trees to capture precious sunlight and thrive, new firms expand by taking advantage of the failure of large incumbents. But when the incumbent fails, there must be a sufficient diversity of small and new entrants who are in a position to take advantage. A long period of stabilisation does its greatest damage by stamping out this diversity and breeding a micro-stable, macro-fragile environment. Just as in ecosystems, “minor species provide a ‘‘reservoir of resilience’’ through their functional similarity to dominant species and their ability to increase in abundance and thus maintain function under ecosystem perturbation or stress”. This deterioration is not evident during the good times when the dominant species, however homogeneous, appear to be performing well. Stabilisation is therefore an almost irreversible path – once the system is sufficiently homogenous, avoiding systemic collapse requires us to put the incumbent fragile players on permanent life support.
As even Marxists such as David Harvey admit, Olsonian special-interest dynamics subvert and work against the interests of the class struggle:
the social forces engaged in shaping how the state–finance nexus works…differ somewhat from the class struggle between capital and labour typically privileged in Marxian theory….there are many issues, varying from tax, tariff, subsidy and both internal and external regulatory policies, where industrial capital and organised labour in specific geographical settings will be in alliance rather than opposition. This happened with the request for a bail-out for the US auto industry in 2008–9. Auto companies and unions sat side by side in the attempt to preserve jobs and save the companies from bankruptcy.
This fleeting and illusory stability that benefits the short-term interests of the currently employed workers in a firm leads to the ultimate loss of bargaining-power and reduced real wage growth in the long run for workers as a class. In the pursuit of stability, the labour class supports those very policies that are most harmful to it in the long run. A regime of Smithian efficiency-seeking i.e. the invisible hand, without Schumpeterian disruption i.e. the invisible foot inevitably leads to a system where capital dominates labour. Employed workers may achieve temporary stability via special-interest politics but the labour class as a whole will not. Creative destruction prevents the long-term buildup of capital interests by presenting a constant threat to the survival of the incumbent rent-earner. In the instability of the individual worker (driven by the instability of their firm’s prospects) lies the resilience of the worker class. Micro-fragility is the key to macro-resilience but this fragility must be felt by all economic agents, labour and capital alike.
In a previous post I argued that in the current environment, the Federal Reserve could buy up the entire stock of government bonds without triggering any incremental inflation. The argument for the ineffectiveness of conventional QE is fairly simple. Government bonds are already safe collateral both in the shadow banking system as well as with the central bank itself. The liquidity preference argument is redundant in differentiating between deposits and an asset that qualifies as safe collateral. Broad money supply is therefore unaffected when such an asset is purchased.
The monetarist objection to this argument is that QE increases the stock of high-powered money and increases the price level to the extent that this increase is perceived as permanent. But in an environment where interest is paid on reserves or deposits with the central bank, the very concept of high-powered money is meaningless and there is no hot potato effect to speak of. Some monetarists argue that we need to enforce a penalty rate on reserves to get rid of excess reserves but small negative rates make little difference to safe-haven flows and large negative rates will lead to people hoarding bank notes.
The other objection is as follows: if the central bank can buy up all the debt then why don’t we do just that and retire all that debt and make the state debt-free? Surely that can’t be right – isn’t such debt monetisation the road to Zimbabwe-like hyperinflation? Intuitively, many commentators interpret QE as a step on the slippery slope of fiscal deficit monetisation but this line of thought is fatally flawed. Inflation comes about from the expected and current monetisation of fiscal deficits, not from the central bank’s purchase of the stock of government debt that has arisen from past fiscal deficits. The persistent high inflation that many emerging market economies are so used to arises from money-printed deficits that are expected to continue well into the future.
So why do the present and future expected fiscal deficits in the US economy not trigger inflation today? One, the present deficits come at a time when the shadow money supply is still contracting. And two, the impact of expected future deficits in the future is muddied thanks to the status of the US Dollar as the reserve currency of the world, a status that has been embellished since the 90s thanks to reserves being used as capital flight and IMF-avoidance insurance by many EM countries (This post by Brett Fiebiger is an excellent explanation of the privileged status enjoyed by the US Dollar). The expectations channel has to deal with too much uncertainty and there are too many scenarios in which the USD may hold its value despite large deficits, especially if the global economy continues to be depressed and demand for safe assets remains elevated. There are no such uncertainties in the case of peripheral economy fiat currencies (e.g. Hungary). To the extent that there is any safe asset demand, it is mostly local and the fact that other global safe assets exist means that the fiscal leeway that peripheral economies possess is limited. In other words, the absence of inflation is not just a matter of the market trusting the US government to take care of its long-term structural deficit problems – uncertainty and the “safe asset” status of the USD greatly diminish the efficacy of the expectations channel.
Amidst the fog of uncertainty and imperfect commitments, concrete steps matter and they matter especially in the midst of a financial crisis. Monetary policy can almost always prevent deflation in the face of a contraction in shadow money supply via the central banks’ lender-of-last-resort facilities. In an economy like 2008-2009, no amount of open-market operations, asset purchases and monetary target commitments can prevent a sharp deflationary contraction in the private shadow money supply unless the lender-of-last-resort facility is utilised. Once the system is stabilised and the possibility of a deflationary contraction has been avoided, monetary policy has very little leeway to create incremental inflation in the absence of fiscal profligacy and shadow banking/private credit expansion except via essentially fiscal actions such as buying private assets, credit guarantees etc. In the present situation where the private household economy is excessively indebted and the private business economy suffers from a savings glut and a persistent investment deficit due to structural malformation, fiscal profligacy is the only short-term option. Correspondingly, no amount of monetary stimulus can prevent a sharp fiscal contraction from causing deflation in the current economic state.
Monetary policy is also not all-powerful in its contractionary role – it has significant but not unlimited leeway to tighten policy in the face of fiscal profligacy or shadow banking expansion. The Indian economy in 1995-1996 illustrates how the Reserve Bank of India (RBI) could control inflation in the face of fiscal profligacy only by crippling the private sector economy. The real rates faced by the private sector shot up and spending ground to a halt. The dilemma faced by the RBI today mirror the problems it faced then – if fiscal indiscipline by the Indian government persists, the RBI cannot possibly bring down inflation to acceptable levels without causing the private sector economy to keel over.
The current privileged status of the US Dollar and the low interest rates and inflation does not imply that long-term fiscal discipline is unimportant. Currently, the demand for safety reduces inflation and the low inflation renders the asset safer – this virtuous positive-feedback cycle can turn vicious if expectation of monetisation is sufficiently large and the mutual-feedback nature of the process means that any such transition will almost certainly be rapid. It is not even clear that the United States is better off than say Hungary in the long run. The United States has much leeway and flexibility than Hungary but if it abuses this privilege, any eventual break will be that much more violent. Borrowing from an old adage, give an economy too much rope and it will hang itself.
As I highlighted in my previous post, the honeymoon period for the SNB in its enforcement of the 1.20 floor on the EURCHF exchange rate is well and truly over. In May, the SNB needed to intervene to the tune of CHF 66 bn to defend the floor. There’s even speculation that the SNB may be forced to implement capital controls or negative interest rates on offshore deposits in the event of a disorderly Greek exit from the Eurozone.
Increasingly, the SNB is caught between a rock and a hard place. Either it can continue to defend the peg and accumulate increasing amounts of foreign exchange reserves on which it faces the prospect of correspondingly increasing losses. Or it can abandon the peg, allow the CHF to appreciate 20-25% and risk deflation and a collapse in exports and GDP. It is not difficult to see why the SNB is being forced to defend the peg – the EUR in the current environment is a risky asset and the CHF is a safe asset. By committing to sell a safe asset at a below-market price, the SNB is subsidising the price of safety. It is no wonder then that this offer finds so many takers when there is a flight to safety.
Some argue that the continued deflation in the Swiss economy allows the SNB to maintain its peg but this argument ignores the fact that it is the continued deflation that also maintains the safe status of the Swiss Franc. Deflation provides the impetus for the safe-haven flows due to which the required intervention by the SNB and the SNB’s risk exposure are that much greater in magnitude. Therefore, if the SNB is eventually forced to abandon the floor, the earlier the better. A prolonged period of deflation punctuated by occasional flights to safety will compel the SNB to accumulate an unsustainable level of foreign exchange reserves to defend the floor. By the same logic, the SNB would obviously prefer that the Eurozone not implode but if it does implode, then it would rather that the Euro implodes sooner rather than later.
So what does the SNB need to do? It needs to engineer an outcome where the market price of the EURCHF moves up and the CHF devalues by itself. The only sustainable way to achieve this is to provide a significant dose of inflation to the Swiss economy and it needs to do so in a manner that does not provide an even larger subsidy to those running away from risk. For example, raising the EURCHF floor by itself only increases the temptation to buy the Franc and at best provides a one-time dose of inflation. The SNB could decide to buy CHF private sector assets but the safe-haven inflows and relatively strong performance of the Swiss economy mean that asset markets, especially housing, are already frothy.
The more sustainable and equitable solution is to simply make the safe asset unsafe by generating the requisite inflation for which money-financed helicopter drops are the best solution. Money-financed fiscal transfers will create inflation, deter the safe-haven inflow and shore up the balance sheet of the Swiss household sector. The robustness of this solution in creating sustainable inflation will not come as a shock to any emerging market central banker or finance minister. The crucial difference between this plan and that implemented by banana republics around the world is that instead of printing money and funnelling it to corrupt government officials we will distribute the money to the masses.
If we analyse monetary policy as a threat strategy, then how do we make sure that the threat is credible? According to Nick Rowe, “The Fed needs to communicate its target clearly. And it needs to threaten to do unlimited amounts of QE for an unlimited amount of time until its target is hit. If that threat is communicated clearly, and believed, the actual amount of QE needed will be negative.” In essence, this is a view that a credible threat will cause market expectations to adjust and negate the need for any actual intervention in markets by the central bank.
The current poster-child for this view is the SNB’s maintenance of a floor on the EURCHF exchange rate at 1.20. The market-expectations story argues that because the SNB has credibly committed to maintaining a floor on EURCHF, it will not need to intervene in the markets at all (See Evan Soltas here and here, Scott Sumner, Matthew Yglesias and Timothy Lee). And indeed the SNB did not need to intervene at all…..that is, until May, when they were required to intervene to the tune of CHF 66 billion within the span of just a month in order to defend the floor from euro-crisis induced safe-haven flows.
Even when the central bank wants to hit a target as transparent and as liquidly traded as an exchange rate, it seems that actual intervention is needed sooner or later. Therefore when the transmission channel between central bank purchase of assets and the target variable is as blurred as it would be in regimes such as NGDP targeting, it is unlikely that the central bank will get away with just waving the magic wand of market expectations.
Many economists and commentators blame the Federal Reserve for the increasingly tepid economic recovery in the United States. For example, Ryan Avent calls the Fed’s unwillingness to further ease monetary policy a “dereliction of duty” and Felix Salmon claims that “we have low bond yields because the Fed has failed to do its job”. Most people assume that the adoption of a higher inflation target (or an NGDP target) and conventional quantitative easing (QE) via government bond purchases will suffice. Milton Friedman, for example, had argued that government bond purchases with “high-powered money” would have dragged Japan out of its recession. But how exactly is more QE supposed to work in an environment when treasury bonds are trading at all-time low yields and banks are awash in excess reserves?
If we analyse monetary policy as a threat strategy, then how do we make sure that the threat is credible? According to Nick Rowe, “The Fed needs to communicate its target clearly. And it needs to threaten to do unlimited amounts of QE for an unlimited amount of time until its target is hit. If that threat is communicated clearly, and believed, the actual amount of QE needed will be negative.” In essence, this is a view that market expectations are sufficient to do the job.
Expectations are a large component of how monetary policy works but expectations only work when there is a clear and credible set of actions that serve as the bazooka(s) to enforce these expectations. In other words, what is it exactly that the central bank threatens to do if the market refuses to react sufficiently to its changed targets? It is easy to identify the nature of the threat when the target variable is simply a market price, e.g. an exchange rate vs another currency (such as the SNB’s enforcement of a minimum EURCHF exchange rate) or an exchange rate vs a commodity (such as the abandoning of the gold standard). But when the target variable is not a market price, the transmission mechanism is nowhere near as simple.
Scott Sumner would implement an NGDP targeting regime in the following manner:
First create an explicit NGDP target. Use level targeting, which means you promise to make up for under- or overshooting. If excess reserves are a problem, get rid of most of them with a penalty rate. Commit to doing QE until various asset prices show (in the view of Fed officials) that NGDP is expected to hit the announced target one or two years out. If necessary buy up all of Planet Earth.
Interest on Reserves
Small negative rates on reserves or deposits held at the central bank are not unusual. But banks can and will pass on this cost to their deposit-holders in the form of negative deposit rates and given the absence of any better liquid and nominally safe investment options, most bank customers will pay this safety premium. For example, when the SNB charged negative rates on offshore deposits denominated in Swiss Franc in the mid-1970s, the move did very little to stem the inflow into the currency.
Significant negative rates are easily evaded as people possess the option to hold cash in the form of bank notes. As SNB Vice-Chairman Jean-Pierre Danthine notes:
With strongly negative interest rates, theory joins practice and seems to lead to a policy of holding onto bank notes (cash) rather than accounts, which destabilises the system.
Quantitative Easing: Government Bonds
Conventional QE can be deconstructed into two components: an exchange of money for treasury-bills and an exchange of treasury-bills for treasury-bonds. The first component has no impact on the market risk position of the T-bill holder for whom deposits and T-bills are synonymous in a zero-rates environment. But it is also irrelevant from the perspective of the banking system unless the rate paid on reserves is significantly negative (which can be evaded by holding bank notes as discussed above).
The second component obviously impacts the market risk position of the economy as a whole. It is widely assumed that by purchasing government bonds, the central bank reduces the duration risk exposure of the market as a whole thus freeing up risk capacity. But for most holders of government bonds (especially pension funds and insurers), duration is not a risk but a hedge. A nominal dollar receivable in 20 years is not always riskier than a nominal dollar receivable today – for those who hold the bond as a hedge for a liability of a nominal dollar payable in 20 years, the dollar receivable today is in fact the riskier holding. More generally the negative beta nature of government bonds means that the central bank increases the risk exposure of the economy when it buys them.
Apart from the market risk impact of QE, we need to examine whether it has any impact on the liquidity position of the private economy. In this respect, neither the first or the second step has any impact for a simple reason – the assets being bought i.e. govt bonds are already safe collateral both in the shadow banking system as well as with the central bank itself. Therefore, any owner of government bonds can freely borrow cash against it. The liquidity preference argument is redundant in differentiating between deposits and an asset that qualifies as safe collateral. Broad money supply is therefore unaffected when such an asset is purchased.
If conventional QE were the only tool in the arsenal, announcing higher targets or NGDP targets achieves very little. The Bank of England and the Federal Reserve could buy up the entire outstanding stock of govt bonds and the impact on inflation or economic growth would be negligible in the current environment.
Credit Easing and More: Private Sector Assets
Many proponents of NGDP targeting would assert that limiting the arsenal of the central bank to simply treasury bonds is inappropriate and that the central bank must be able to purchase private sector assets (bonds, equities) or as Scott Sumner exhorts above “If necessary buy up all of Planet Earth”. There is no denying the fact that by buying up all of Planet Earth, any central bank can create inflation. But when the assets bought are already liquid and market conditions are not distressed, buying of private assets creates inflation only by increasing the price and reducing the yield of those assets i.e. a wealth transfer from the central bank to the chosen asset-holders. As with quantitative easing through government bond purchases, the inability to enforce adequate penalties on reserves nullifies any potential “hot potato” effect.
Bernanke himself has noted that the liquidity facility interventions during the 2008-2009 crisis and QE1 were focused on reducing private market credit spreads and improving the functioning of private credit markets at a time when the market for many private sector assets was under significant stress and liquidity premiums were high. The current situation is not even remotely comparable – yields on private credit instruments are at relatively elevated levels compared to historical median spreads but the difference in absolute terms is only about 50 bps on investment-grade credit (see table below) as compared to much higher levels (at least 300-40 bps on investment grade) during the 2008-2009 crisis.
A quantitative easing program focused on purchasing private sector assets is essentially a fiscal program in monetary disguise and is not even remotely neutral in its impact on income distribution and economic activity. Even if the central bank buys a broad index of bonds or equities, such a program is by definition a transfer of wealth towards asset-holders and regressive in nature (financial assets are largely held by the rich). The very act of making private sector assets “safe” is a transfer of wealth from the taxpayer to some of the richest people in our society. The explicit nature of the central banks’ stabilisation commitment means that the rent extracted from the commitment increases over time as more and more economic actors align their portfolios to hold the stabilised and protected assets.
Such a program is also biased towards incumbent firms and against new firms. The assumption that an increase in the price of incumbent firms’ stock/bond price will flow through to lending and investment in new businesses is unjustified due to the significantly more uncertain nature of new business lending/investment. This trend has been exacerbated since the crisis and the bond market is increasingly biased towards the largest, most liquid issuers. Even more damaging, any long-term macroeconomic stabilisation program that commits to purchasing and supporting macro-risky assets will incentivise economic actors to take on macro risk and shed idiosyncratic risk. Idiosyncratic risk-taking is the lifeblood of innovation in any economy.
In other words, QE is not sufficient to hit any desired inflation/NGDP target unless it is expanded to include private sector assets. If it is expanded to include private sector assets, it will exacerbate the descent into an unequal, crony capitalist, financialised and innovatively stagnant economy that started during the Greenspan/Bernanke put era.
Removing the zero-bound
One way of getting around the zero-bound on interest rates is to simply abolish or tax bank note holdings as Willem Buiter has recommended many times:
The existence of bank notes or currency, which is an irredeemable ‘liability’ of the central bank – bearer bonds with a zero nominal interest rate – sets a lower bound (probably at something just below 0%) on central banks’ official policy rates.
The obvious solutions are: (1) abolishing currency completely and moving to E-money on which negative interest rates can be paid as easily as zero or positive rates; (2) taxing holdings of bank notes (a solution first proposed by Gesell (1916) and also advocated by Irving Fisher (1933)) or (3) ending the fixed exchange rate between currency and central bank reserves (which, like all deposits, can carry negative nominal interest rates as easily as positive nominal interest rates, a solution due to Eisler (1932)).
I’ve advocated many times on this blog that monetary-fiscal hybrid policies such as money-financed helicopter drops to individuals should be established as the primary tool of macroeconomic stabilisation. In this manner, inflation/NGDP targets can be achieved in a close-to-neutral manner that minimises rent extraction. My preference for fiscal-monetary helicopter drops over negative interest-rates is primarily driven by financial stability considerations. There is ample evidence that even low interest rates contribute to financial instability.
There’s a deep hypocrisy at the heart of the macro-stabilised era. Every policy of stabilisation is implemented in a manner that only a select few (typically corporate entities) can access with an implicit assumption that the impact will trickle-down to the rest of the economy. Central-banking since the Great Moderation has suffered from an unwarranted focus on asset prices driven by an implicit assumption that changes in asset prices are the best way to influence the macroeconomy. Instead doctrines such as the Greenspan Put have exacerbated inequality and cronyism and promoted asset price inflation over wage inflation. The single biggest misconception about the macro policy debate is the notion that monetary policy is neutral or more consistent with a free market and fiscal policy is somehow socialist and interventionist. A program of simple fiscal transfers to individuals can be more neutral than any monetary policy instrument and realigns macroeconomic stabilisation away from the classes and towards the masses.