Teaching

Monetary Policy

In August 2007, some chilling news spread around the corridors of Threadneedle Street: Northern Rock was failing. Since that day, almost 15 years ago, central banking has been revolutionised. Once a quiet, pseudo-academic calling, monetary policymakers have become superstars, the heroes and heroines we rely on to do whatever it takes to rescue ailing economies. Central banks control cash, and in modern economic policy, cash is king

That central banks are now seen as the only game in town would have surprised classical economists, and the founders of modern economics alike. For Hume and Ricardo, Keynes and Friedman monetary policy was a limited tool, often thought of as a ‘string’: you could pull on it to slow an economy down but could not reliably push on it to speed activity up. Today things are different: money is our accelerator, and rarely a brake.

Central banks are staffed by some of the brightest economists, but there are reasons for concern about the challenges they face. If the past 15 years have been ones of turmoil, I believe the next 15 years will be an even sterner test, for the following three reasons:

• The transmission mechanism. Starting in 2005, rising concerns about the extent to which central bank policy decisions (interest rates, and QE) are transmitted to the economy.

• New forms of money. The emergence of cryptocurrencies, and in particular ‘stablecoins’ such as Tether. These may come to provide an alternative to central bank reserves.

• Threats to the framework. The independence of central banks is threatened by the extent to which they have become the main buyers of government debt.

Here I focus on the first of these. The second is a bubbling concern but can be avoided by CBs issuing their own coins, which I believe is inevitable (China is already doing it). The third is important it will be overcome: the problem is recognised and influential voices have called for a new constitution for central banking, which I think will (in a few year) happen. But the first gets to the very heart of central bank’s power. It is the existential question. Is the transmission mechanism weakening? And if it is, how are central banks going to influence the economy as we chart a course to 2030?

The road to inflation targeting: UK experience

To understand where we are, we need to understand where we have been. Post-War monetary policy in the UK can be split into seven distinct periods:

• 1948 – 1971. The Bretton-Woods system. Fixed but adjustable exchange rates.

• 1971 – 1976. Floating rates, no nominal anchor.

• 1976 – 1987. Monetary aggregate targets.

• 1987 – 1992. Exchange rate targets. DM shadowing and the ERM

• 1992 – 1997. Inflation targeting. (IT)

• 1997 – 2008. Inflation targeting + independence. (IT&I). Tool: interest rates.

• 2009 – 2021. IT&I. Tools: QE, forward guidance.

These periods are demarked by vertical lines on the chart of UK inflation below.

The code to draw this chart is here. The data is here.

The volatility of the past is clear from this chart. Inflation was over 25% in the UK in the early 1970s following the first oil shock (in 1973) and went over 20% following the second (in 1979). This inflation performance was worse than neighbouring countries, including Germany. Alongside volatile prices the UK experienced huge swings in interest rates, as the next chart shows.

This is taken from the Bank of England’s database. The code for the chart is here.

In both charts, 1992 stands out as a year in which things changed. In the preceding years this, the UK had been tied—formally or informally—to decisions made by the Bundesbank: first, as the deutschmark was ‘shadowed’ in the late 1980s, and then withing the Exchange Rate Mechanism. Maintaining the value of sterling meant interest rates had to be high, playing a role in the UK’s 1990 recession.

After crashing out of the ERM in 1992 the UK switched to inflation targeting. Things improved almost immediately: inflation dropped, and interest rates were no longer volatile, instead rising and dipping gently as the Bank of England sought to subtly guide the economy. The move to an independent central bank in 1997 helped by anchoring inflation expectations further. This new system was based on research that had clarified the inflation-unemployment-expectations trade-offs. This work, by Bill Phillips, Robert Solow and Paul Samuelson, Milton Friedman, and Finn Kydland with Ed Prescott, was ground breaking. [A full reading list with all links is provided below].

These academic foundations were impeccable, and the system worked, with inflation in the UK not only falling, but becoming much more stable. The costs of inflation—whether anticipated or unexpected—evaporated.

Of course, monetary policy was not the sole cause. Across the world inflation fell and is now much lower than it was in the 1970s. (The chart below has the inflation rate in 2020 for hundreds of countries – you can see the names and inflation rates by hovering over it with your mouse.) The reduction in inflation is complex, due to many trends including more stable commodity markets, trade, demographics and technology. Despite this some countries still suffer from much higher inflation, including some many of those with the highest populations. For the average person on the planet, and for the very poorest, inflation is still a problem.

The shifting Phillips Curve

One relationship at the heart of central banks modelling and thinking is the Phillips Curve. Phillips (for the UK) and then Solow/Samuelson (for the US) found a clear relationship: lower unemployment meant higher wage inflation. Over time evidence on this has changed: since the 1960s—particularly in the US and UK—this curve has seemingly flattened.

The chart below plots 260 years of inflation-unemployment data for the UK. It uses the Bank of England’s Millennium of Data together with the latest numbers from the ONS. By clicking on the legend you can select the data for 1860-1960, the period Phillips was interested in. Here the relationship is strong. For the other periods the relationship is harder to see.

This is not to say the Phillips Curve is irrelevant or wrong, just that macro models need to evolve as economic relationships do. Overall, evidence here is mixed and it is an important area of economic research. In the US, regional (state by state) Phillips curves show a flattening, akin the to data above. But you can still easily find a Phillips type relationship of you plot cross country data, as I have done using 40 years of WEO data from the IMF in the chart below.

Inflation targeting, and the transmission mechanism

Returning to the inflation targeting story for the UK, we can see in the chart for Bank Rate about that between 1992 to 2008 interest rates were used actively. There are five distinct tightening and loosening periods. The bank was using the interest rate as a rudder, gently guiding the UK economy to a point—the inflation target—on the horizon. The line wiggles along, with the BoE (and Central Banks across the world) generally getting to where the wanted to go. The success of the early adopters, including New Zealand, Canada and Australia mean that inflation targeting became a popular policy.

At the heart of this new model lay an important relationship: the “transmission mechanism” which explained how changes in interest rates fed through to changes in inflation. A complex idea and the topic of a huge amount of discussion and research in central banks it has four main channels.

• Market interest rates. Central banks rates feed through to the rates paid by households and firms. This stimulates or slows investment and consumption.

• Asset prices. Interest rates change the present value of assets, leading to collateral and wealth effects that can stimulate investment and consumption.

• Exchange rates. In a small open economy, a cut in the policy rate will lead to a depreciating, generating imported inflation and improving the trade balance.

• Expectations. Communication of a clear and stimulative policy path (i.e. rates are going to be held low until inflation rises) can spur activity and inflation.

A diagram, which has become a staple, used in many subsequent publications and textbooks was set out by the Bank of England in 1999.

The GFC and the rise of QE

The crash of 2008 changed everything in central banking. With interest rates pinned to the floor central banks looked to Japan, which had been in a similar situation ever since its 1990 property bubble and recession. By 2008, as the chart below shows, Japan had already had 12 years of ultra-low interest rates.

The way to stimulate the economy in Japan was with quantitative easing (QE). The Bank, Federal Reserve and ECB followed the BoJ and started buying up assets in large amounts. Central banks had lost their main policy tool and were now relying on another.

QE can seem a confusing way to try to simulate the economy, but in many ways the idea is the same as with interest rates changes. Central banks buy up government bonds to push up their prices and lower market interest rates. This also tends to devalue the currency. The ‘transmission mechanism’ in other works, is very similar. This means that for QE to work and for there being any hope of getting back to using interest rates as an economic rudder, the transmission mechanism set out in the diagram above needs to work.

Is the mechanism broken?

The problem is that the previously clear routes for transmission of a central bank decision to the economy have become muddied.

Interest rates

Concern over the link between official central bank rates and market rates pre-dates the financial crisis. In 2005 Alan Greenspan had outlined a ‘conundrum’ –the Federal Funds rate was increased, yet market rates—in particular the yield on US government debt—did not respond. Whatever force was causing this conundrum, it meant that market interest rates were lower than most powerful banking the world wanted them to be.

Following the 2008 crash the reverse happened: interest rates stayed stubbornly high. In the UK, the main policy rate—Bank Rate—was set to just 0.5% but mortgage rates failed to react, as the chart below shows. Spreads—the difference between policy and market rates—were higher, and remained so. Two years after rates had been slashed, new UK residential fixed-rate mortgages still had interest rates of over 4%. Both periods show central banks’ ability to control market rates can change over time, and is not guaranteed.

In addition to a high cost of credit, concerns emerged about the quantity of credit supplied, particular to companies. Lending to firms began to contract, as the chart below shows. While credit supply may have been too loose in the early 2000s, the scale of the credit withdrawal has been surprising. At the 2009 peak, £495 billion in credit was provided to UK firms. By mid-2015 that had fallen to £343 billion, a 30% reduction. In early 2021, more than a decade on, credit supply by UK banks to UK firms has still not fully recovered.

Asset prices

The evidence suggest that this part of the transmission mechanism still works, given a narrow set of aims. Quantitative easing has driven up asset prices in, lowering yields and supporting GDP, inflation and employment. [See papers in reference list on QE].

Seen through a wider lens, problems occur. Public communication is a foundational element of the inflation targeting & independence framework adopted in the UK. Success relies on clarity, and on public confidence and acceptance of the system. There are at least three problems:

• Inequality. Rising concerns that QE exacerbates pre-exiting imbalances in the distribution of wealth.

• Pollution. Extended QE is some central banks to buy corporate bonds. The companies with suitable large-scale bond issuance are often engaged in heavy manufacturing, utilities, or agriculture. CBs are at risk of being criticized for a role in supporting polluting economic activity.

• Independence. Rising concerns about the close relationship between finance ministries and central banks that can emerge when monetary policymakers are the main buyers of government debt.

These risks do not make QE a bad tool: all policies have downsides. But QE does creates a thorny communication issue for central banks. In the 1980s this may not have mattered, but the new model of central banking is one in which public understanding and acceptance of central banks’ aims is vital.

Exchange rates

The penultimate risks is a simple one, and can be made succinctly with pair of charts. The UK has experienced two large depreciations in the past 15 years—in 2008, and in late 2016—as the charts below shows.

Formal macro models and informal narratives suggest an improvement in the trade balance with such a step change in the exchange rate. The chart below shows what actually happened – very little. The exchange rate channel, which is supposed to boost output via the trade balance, is an uncertain one.

Expectations

At present, the final channel seems to be holding up: central bank announcements clearly influence expectations of growth and inflation. This is relieving. But this channel, which is given the same weight in the flow diagrams set out above, is surely contingent on or junior to the others. Central banks can influence our expectations, but that is because we know they can influence things that matter to us: interest rates, asset prices and the exchange rate. If these three channels are damaged, the expectations channel will weaken too. And if that happens, our central banks may end up far weaker than we imagine, and our continual reliance on monetary policy to boost ailing economies may turn out to be unwise.

Richard Davies, 29th March 2021

Sources: data and papers

Data sources

Here are some links I use in research and writing. The ONS is the main source, but for the UK many vital files are hidden deep within websites of various agencies and regulators. This includes:

· BoE. Bank Stats tables. https://www.bankofengland.co.uk/statistics/tables

· BoE. Interactive Database. (Old version, still active, is much more intuitive than the new version). http://www.bankofengland.co.uk/boeapps/iadb/ . Useful series:

o LPMBC55 – HH lending.

o LPMZ5MO –PNFC lending.

· FCA. Mortgage lending statistics. https://www.fca.org.uk/data/mortgage-lending-statistics

For the US the main source is FRED. Other useful sources:

· The Fed’s balance sheet releases. https://www.federalreserve.gov/releases/h41/current/

What is money? Where does it come from?

Karl Menger (EJ, 1892), On the Origin of Money.

RA Radford (Economica, 1945), The Economic Organization of a POW Camp.

Charles Goodhart (EJPE, 1988), The two concepts of money.

Richard Davies (Guardian, 2019), Prison Currencies.

Richard Davies (Economist, 2012), On the Origin of Specie.

Monetary neutrality, non-neutrality, the role of policy