A long-run view on monetary policy - Part 3: Causes and consequences of the SNB's large balance sheet

This is Part 3 of a series of articles on Swiss monetary policy I wrote jointly with Simon Schmid for Republik. They kindly agreed that I can publish an english version on my blog. Enjoy:

The balance sheet of the Swiss National Bank is currently about one fourth larger than annual nominal GDP. This article is a monetarist journey to show the causes and consequences of the exceptionally large balance sheet.

Missing monetary policy traction

The most important reason why the SNB's balance sheet grew so much since the financial crisis were foreign exchange interventions: The SNB bought securities in euro, US dollar, and other foreign currencies currently worth about 750 billion. We can describe these interventions in two ways

• In terms of a balance sheet expansion: The SNB paid for the foreign securities with newly created Swiss francs. This increases the liability side of the balance sheet while the foreign securities increase the asset side of the balance sheet (Simon Schmid nicely described how this works in detail)
• In terms of an increase of the money stock: The SNB increased the amount of Swiss francs in the financial system by increasing the monetary base (that is reserve accounts of commercial banks at the SNB and banknotes in circulation).

If a central bank increases the money stock, many commentators write that the central bank ramps up the the printing press and pumps liquidity into the market. These metaphors are not wrong, but we have to be careful when judging whether monetary policy is expansionary as a result.

Central bank balance sheets fluctuate strongly over time. In crises they often expand, but it takes a long time to reduce them to their previous levels (compared to nominal GDP). The size of the money stock is actually an imperfect indicator of how expansionary monetary policy is: Sometimes, monetary policy looses traction so that an increase in the money stock has little effect on the economy. To illustrate this, think about the following example:

• You try to drive up a hill on a snowy street with summer tires. Although you floor the gas pedal the car hardly advances because your tires lack traction.
• If you use winter tires, or if there is no snow on the street, you have good traction and the car advances. Flooring the gas pedal will immediately accelerate your car.

We observe a similar phenomenon in monetary policy. The central bank can ramp up the printing press; if monetary traction is low, however, this will have little impact on the economy. 

Milton Friedman and Anna Schwartz described how monetary traction changes over time in their monetary history of the United States. The impact of monetary policy is low in recessions, and in particular, during severe financial crises. In such crises, the US central bank had to compensate money creation by commercial banks because they were not willing or able to provide sufficient liquidity for the economy. If the central bank hesitated to do so (as for example during the Great Depression) they documented severe economic downturns.

The loss of traction is important to understand Swiss monetary policy, in particular during the last financial crisis. A measure of the traction is the velocity of money. It approximately shows how many times one Swiss franc created by the Swiss National Bank was used in a transaction. If this measure is high, one Franc will be used for more transactions, if it is low for fewer transactions.

Figure 1: Loss of traction

Ratio of nominal GDP to the monetary base (velocity of money). Sources: Historical Statistics Switzerland Online (HSSO), Swiss Federal Statistical Office (SFSO), SECO, Historical time series (SNB). 

We observe two strong declines: during the Great Depression and during the recent financial crisis. This confirms that monetary policy traction usually collapses during severe crises. As a consequence, a given increase in the money stock is associated with a smaller increase in economic activity. 

Apparently, monetary policy traction changes over time, as does the traction of car tires. For car tires, traction depends on the temperature, the surface, the type of tire. For monetary policy, traction depends on the health of the banking sector, the state of the business cycle, the level of interest rates, and the monetary policy strategy. 

Last week we described that the level of interest rates steadily declined over the past 40 years. This decline is an important reason for the missing traction of monetary policy. Intuitively, banks create little new money because interesting investment opportunities are missing. In addition, they bunker most of the newly created central bank money for the same reason. Paul Krugman formalized this idea 20 years ago: when interest rates are zero, a substantial increase in the monetary base will not bolster economic activity or lead to higher inflation. The collapse in the velocity of money during the financial crisis actually resembles such a situation.

If monetary policy traction is currently so low: why did the SNB expand its balance sheet? Did the SNB print money for now good reason?

How various countries managed the crisis

The answer to the last question is: no. Or to use the metaphor: the monetary wheels of the SNB were spinning in the crisis. But at least the economy did not roll backwards down the hill.

What we actually have to ask is the following: What would have happened if the SNB would not have expanded the money stock and its balance sheet? Such what-if-questions make macroeconomics interesting but challenging. We will still attempt to give an answer comparing various countries during and after the financial crisis.

The following figure shows real GDP indexed to 100 in the fourth quarter 2007 (just before the crisis). We see that Switzerland fared relatively well in particular during the first phase of the crisis. Real GDP fell much less in 2008 and returned quickly to its pre-crisis level. 

Figure 2: Switzerland fared well during and after the crisis

Real GDP indexed to 100 in Q4 2007. Sources: Eurostat, U.S. Bureau of Economic Analysis.

In addition, the next figure shows that the money stock increased more quickly than in other countries; especially during the first phase of the crisis. We show the money stock M1, which includes the monetary base controlled by the central bank as well as deposits created by commercial banks, mainly because it is available for many countries from the OECD. So the Swiss money stock increased substantially. One reason for this expansion are the foreign exchange interventions. In addition, the Swiss banking sector was in a relatively good state compared to, for example, the euro zone.

Figure 3: The money stock expanded more quickly in Switzerland

Money stock M1 indexed to 100 in Q4 2007. M1 includes the monetary base as well as short-term deposits at commercial banks. Source: OECD.

The Swiss example suggests that the expansion of the money stock helped mitigating the impact of the financial crisis on the real economy. We can reach the same conclusion when comparing other countries:

• United States and the euro zone: In 2009, the European Central Bank increased interest rates, which stopped the increase in the money stock. Meanwhile, the Federal Reserve kept US interest rates low and conducted three rounds of large-scale asset purchases (Quantitative Easing). As a result, the money stock in the two monetary areas diverged. At the same time, real GDP developed much more favourably in the US than in the euro zone.
• Sweden and Denmark: Comparing Sweden and Denmark is particularly interesting because Denmark pegs its currency to the euro (that is it implements the same monetary policy), but it does not have the same structural problems as some euro zone countries. We see that real GDP in Denmark developed similarly as in the euro zone. Sweden, with an independent monetary policy, expanded the money stock more strongly and the real economy also developed more favourably. This suggests that monetary policy in the euro zone was too tight even for countries with sound fiscal policy.
• Czech Republic: Up to 2013 the Czech economy was in a relatively bad state. As a result, the central bank lowered interest rates to zero and announced, similar as the SNB, an exchange rate target vis-à-vis the euro. This policy led to a stronger increase in the money stock. As of 2014, the economy started to expand more strongly as well.

These examples all suggest that differences in monetary policy played a role for how different economies managed the crisis. If the SNB would not have expanded its balance sheet, the crisis would have had a stronger impact on the Swiss economy. This is also the implication of a study by the IMF suggesting that, without exchange rate interventions, the Swiss franc would have permanently appreciated to 0.8 to the euro. 

Therefore, the large balance sheet expansion was a necessary reaction to a severe financial crisis and helped to stabilize the economy. Only looking at the money stock, however, does not explain why the SNB had to expand its balance sheet much more strongly than other countries. To explain this, we can once more look at the traction of monetary policy (the velocity of money). 

Figure 4: Switzerland's large loss of traction

Velocity of M1 indexed to 100 in Q4 2007. Source OECD (own calculations)

We see that in Switzerland velocity declined more strongly than in other countries. One reason for this is that the level of interest rates is particularly low (as discussed last week). Of course, we have to be careful when interpreting these charts because the strong increase in the money stock in Switzerland is also one reason for why the velocity declined more strongly than in other countries (increasing the money stock lowers interest rates and therefore lowers velocity). 

But the analysis also shows that we cannot judge how expansionary monetary policy actually is if we do not know its traction. As in the car metaphor: we have to know the position of the gas pedal and the traction of the tires to know whether the car will advance.

We have seen that most central banks lost traction over the past ten years. The large expansion of their balance sheets is a symptom of the missing traction. This is also one take-away of a research paper published in 2014. A substantial share of the balance sheet expansion can be explained by financial market uncertainty and other shocks during the financial crisis. Unexpected monetary policy shocks explain only a small fraction of the balance sheet expansion. Therefore, we should be careful to interpret the large balance sheets as a particularly expansionary monetary policy.

Risks and side-effects

Still, some ask how quickly the SNB will now reduce its balance sheet given that the economy is doing relatively well. Others suggest that the balance sheet is close to its maximum. Some fear inflation and others argue that the SNB will likely expand its balance sheet in the next crisis.

So, how problematic is the large balance sheet? Let's have a look at the risk and side-effects:

• One side-effect are discussions about the composition of the SNB's portfolio and the creation of a sovereign wealth fund. These discussions are largely political in nature and not very important for monetary policy. The main problem from a monetary perspective is that, if the SNB would like to reduce the money stock after transferring foreign assets to a wealth fund, it would have to issue debt certificates (SNB Bills). This could lead to losses for the SNB while the wealth fund makes profits. 
• Another side-effect is that the book losses and profits of the SNB fluctuate more strongly. This is not a monetary policy problem as such but actually a problem of management. There needs to be an arrangement that smooths the distribution of profits to the cantons.
• One important risk, in my view, could be that the size of the balance sheet will be judged to be too large according to some criterion. Accordingly, the SNB may hesitate to expand the money stock during the next severe downturn. This is mainly a problem of central bank independence. From an economic perspective, the SNB could expand its balance sheet as much as necessary. 

From a technical perspective, the large balance sheet is therefore not an issue. If we take into account the political environment, however, the large balance sheet could be a constraint. Should the SNB therefore reduce its balance sheet as quickly as possible?

Central banks should achieve their goals efficiently, that is, with as little interventions as possible. Simply asking for a smaller balance sheet, however, is not a solution. In the current low interest rate environment the traction of monetary policy is still hampered. Taking the foot off the pedal and letting the car slide backwards down the hill is therefore not a particularly sensible strategy.

It would be better to exchange summer tires with winter tires. This means we have to think about whether there are alternative monetary policy strategies that would allow to reach the same goals (price stability while taking account the economic situation) with a smaller balance sheet. For Switzerland, this is even more important than for other countries because the SNB's traction is particularly low.

In the coming weeks, we will see that such strategies indeed exist. But to actually implement them in reality is much more difficult than to show that they work in theory.