In order to identify the responses central banks and fiscal policy leaders can take if a recession were to occur, it is worth characterising two major types of recessions. The first is a typical boom-and-bust recession, caused by downturns in the business cycle. This could be because of demand-side shocks as seen in the 1990s, supply-side shocks in the oil crisis of 1973, or indeed both, as was seen in the 1980s. The second is what Richard Koo describes as a balance sheet recession, whereby there is a collapse in asset prices, while firms are stuck with the loans for those assets. Facing damaged balance sheets, they turn from maximising profit to minimising debt, causing a lack of spending as in Japan’s Lost Decade.
Boom-and-bust recessions can be caused by insufficient aggregate demand. Expansionary monetary policy can be used to counteract this. Central banks attempt to lower interest rates, such that borrowing is cheaper and saving is less profitable, thus encouraging more consumption and stimulating demand. They can lower the base rate at which they lend to commercial banks, and when it is cheaper for commercial banks to borrow, they lower the interest rates they charge borrowers.
However, it is increasingly acknowledged that central bank rates will be permanently low, with it being 0.75% in the UK and 2.25-2.50% in the USA, much lower than the 5.75% and 5.25% respectively in mid-2007. What this means is that central banks are less able to lower interest rates, and the capacity for monetary policy is limited. This zero lower bound exists because if central banks lowered their rates to a negative interest rate and commercial banks passed that on, people could take out cash, which pays 0% interest but is better than losing money.
Hence, what is required is an expansion of the unconventional monetary tools that were developed in the 2008 crisis, which provided an additional 191% of GDP growth on top of the fiscal instruments. For example, quantitative easing was used in 2008, whereby central banks bought government securities. By depositing payments into the commercial bank accounts of the seller, it increases the amount of money commercial banks can lend. Quantitative easing lowers the yields on government bonds as they become more expensive, which incentivises the former owners of bonds to purchase other financial assets and increases their value, and this wealth effect increases the propensity of asset holders to spend.
However, aggressive open market operations require central banks to increase liabilities on their balance sheet. This is particularly dangerous if riskier financial instruments are bought alongside government bonds, as in the US with $1.6 trillion of MBSs currently owned by the Federal Reserve. Indeed, its total assets quadrupled between September 2008 and January 2014, and its board of governors suggested that ‘the longer-run size’ of its balance sheet would be ‘larger than before the crisis’, making future QE more difficult. Its effects are mitigated by most of the new assets bought being owned by the wealthiest, who have a lower marginal propensity to spend. Indeed, the Bank of England reported that the top 5% of households owned 40% of purchased assets, and so their first £375 billion round of quantitative easing stimulated a mere £39 billion in extra spending.
The issues suggest that monetary policy will play a more limited role in facing future recessions. A way to mitigate these flaws is to utilise the costs of holding cash to have negative interest rates. For example, storing money is expensive, and it is a hassle to go to the bank regularly to withdraw cash. Estimates suggest that these costs of holding cash allow the lower bound to go to -0.5/-0.75%, allowing the negative interest rate policies Japan and Denmark have tried. These costs can be increased by phasing out physical cash, or at least high-denomination bank notes. Another solution is a parallel virtual currency, where prices would be in the units of that currency, with an exchange rate between it and paper cash. Given this has already been tried before successfully with the real/URV in Brazil, it is practically feasible. Nonetheless these are less straightforward, and so it is likely that fiscal stimulus, where governments spend money to boost demand or cut taxes to incentivise individuals and corporations to spend more, will play a key role in the next recession to increase demand.
Automatic fiscal stabilisers kick in as tax receipts dry up and welfare payments increase in a recession, but additional discretionary policies such as the 2008 Recovery Act in the USA are necessary. This stimulates demand and has a multiplier effect where initial spending leads to increased income and greater consumption, and parts of the Recovery Act had up to a 2.5x multiplier. However, fiscal spending depends on the ability for governments to borrow and pay interest, and the willingness of lenders to lend is influenced by the government debt-to-GDP ratio. In the US, it is estimated that it will rise to 93% by 2029, the highest since World War Two, leaving much less fiscal room than the 35.2% before the crisis. Large stimuli such as infrastructure projects take a long time, and the time lag limits the countercyclical effects, especially if it crowds out private investment by raising interest rates on loans.
Given these constraints, one alternative fiscal stimulus could be reducing payroll tax contributions, a much faster response that also increases employment. This would be better at increasing spending compared to an income tax reduction, since the cap on payroll tax means it does not disproportionately help the wealthy who have easier access to credit and a lower marginal propensity to spend. As employers pay into this, there is less of a tax penalty on them for creating and maintaining jobs, as well as on workers for entering jobs, and cutting payroll tax by half could increase employment by 4 million jobs.
Supply-side shocks of a boom-and-bust recession are much harder to combat, as little can be done without waiting for capacity-increasing projects to finish or for the shock to dissipate. One way to soften the blow could be for central banks to target nominal GDP instead of price level. Currently, when prices rise from such a shock, interest rates must be increased to maintain the inflation target and central bank credibility, hurting output. A nGDP target allows central banks to use expansionary monetary policy, spreading the adverse impacts between real growth and the price level such that neither suffers too much.
A balance sheet recession has similar demand-side issues, but importantly, those with negative equity want to deleverage rather than borrow. As such, neither lowering the interest rate nor providing liquidity to banks will work, since banks are likely to use it to improve their balance sheets rather than lend it out, and deposits are withdrawn faster than central banks can expand the money supply. For example, Japan increased its monetary base by over 60% between 2001 and 2006, and yet its money supply barely increased, while lending went down. The deleveraging means the economy is losing demand in the form of savings and debt repayments, causing a deflationary spiral as seen in the Great Depression, with a 46% drop in GDP between 1929 and 1933. This is where fiscal policy is most effective, with government borrowing replacing private debt and preventing banks assets from contracting. The lack of private spending means crowding out is less problematic. Indeed, this was seen in Japan after the 1990 bubble, where corporations and households saved or payed down debts at 10% of GDP, but this was compensated for by ¥460 trillion in public spending up till 2005. Compared to an optimistic counterfactual where it recovered to 1985 levels without intervention, this policy provided over ¥2,000 trillion in consumption, a bargain given its costs.
Although the large government borrowing required is a risk, this is mitigated because injections, coupled with the multiplier effect and accelerator effect (of increased capital spending), will increase the size of the economy. This makes it easier to pay off the debt, especially since there will be greater tax revenue at the same tax rates, lower costs of welfare and inflation that eats away at the costs of debt payments.
Ultimately, it would be impossible to comprehensively catalogue all the possible risk factors and ways a recession could occur. However, when faced with these two most common types of recessions, monetary policy is increasingly insufficient, and even conventional fiscal policy needs updating. Instead, central banks and fiscal policy leaders will need to use unconventional tools, such as the use of electronic money, nGDP targets, subsidies for national insurance or an overwhelming amount of government spending to combat a liquidity trap, as to combat the next recession.
Monetary Policy Committee. Inflation Report: February 2019. London: Bank of England, 2019.
Bank of England. ‘Official Bank Rate History’. 2019. www.bankofengland.co.uk/boeapps/database/Bank-Rate.asp
Bils, Mark and Klenow‚ Pete. ‘Fiscal Stimulus in the Form of Lower Payroll Taxes’. SIEPER Policy Brief (January 2009).
Blinder, Alan and Zandi, Mark. How the Great Recession was brought to an end. New York: Moody’s Analytics, 2010.
Board of Governors of the Federal Reserve System. Federal Reserve Press Release. Washington DC: Federal Reserve, 2007.
Board of Governors of the Federal Reserve System. Federal Reserve Press Release. Washington DC: Federal Reserve, 2019.
Board of Governors of the Federal Reserve System. Federal Reserve Statistical Release. Washington DC: Federal Reserve, 2019.
Board of Governors of the Federal Reserve System. Monetary Policy Report. Washington DC: Federal Reserve, 2019.
Broadbent, Ben. Treasury Select Committee Inquiry: The Effectiveness and Impact of post-2008 UK Monetary Policy. London: Bank of England, 2017.
Congressional Budget Office. Estimated Impact of the American Recovery and Reinvestment Act on Employment and Economic Output in 2014. Washington DC: Congress of the United States, 2015.
Congressional Budget Office. The Budget and Economic Outlook: 2019 to 2029. Washington DC: Congress of the United States, 2019.
Federal Reserve. ‘Recent Balance Sheet Trends’. 2019. www.federalreserve.gov/monetarypolicy/bst_recenttrends.htm
Koo, Richard. ‘The World in Balance Sheet Recession: Causes, Cure and Politics’. Real World Economics Review, no. 58 (December 2011).
Monetary Policy Committee. Quarterly Bulletin 2012 Q3. London: Bank of England, 2012
Rogoff, Kenneth. ‘Costs and benefits to phasing out paper currency’. NBER Working Paper Series, no. 20126 (May 2014).