Keynesian economics
Last updated
Was this helpful?
Last updated
Was this helpful?
In the long run, we are all dead - John Maynard Keynes, 1923
The problem of determining the best policy response to short term variations in unemployment is an open challenge in the field of economics and the opinion of best practices has evolved over time. This section summarizes concept theory of counter-cyclical economic policy response from Core Economics Chp 15 and the latest consensus of best practice from OECD at the time of the global financial crisis of 2008-2010, and IMF for the response to the global pandemic COVID-19 of 2020 (OECD, 2010 ; IMF, 2020). Singapore has successfully utilized fiscal stimulus in past recessions in 1985 (Bercuson, IMF, 1995), Asian financial crisis 1997 global financial crisis in 2009 (MAS, 2011) and recently for the pandemic response (Lee, CNBC, 2020). The term “counter-cyclical” policy derives its name from the assertions that there are “natural” business cycle dynamics in a market economy and assuming Okun’s law that either unemployment or GDP output can be used to measure success or failure. The challenge is twofold, whether the particular policy intervention has the desired effect, and second whether there would be long term consequences that would either diminish general welfare and or reduce the potency of future policy interventions.
Keynes recognized that households have daily material needs for food, shelter and cannot wait for long term structural adjustments to work themselves out. The problem of managing short term periods of unemployment is the primary interest of Keynesian aggregate demand management policy and explained in Chapter 14 of Core Economics. Labor productivity and real wages synchronize only in the long run, but do not always move in sync in the short run. A “recession” is defined as consecutive quarterly negative change in GDP. An increase in real wages faster than labor productivity growth can create inflationary pressure. Inflationary pressure reduces unemployment. In visa versa real wages dropping slower than productivity are associated with depressed conditions, falling inflation and rising unemployment. The concept of the unemployment-inflation trade-off is known as the Phillips curve. While in general there is no fixed Phillips curve across countries and long periods of time, at any point in time in a given country these two measures behave as though they are sliding on a single fixed Phillips curve (Core Economics Chp 15, 2015).
There are two types of response strategies to short term rise in unemployment - supply side and demand side. The empirical evidence for demand side strategy is the relationship between GDP and unemployment in the short run. Figure A.1 illustrates Okun’s law - the observation that changes in quarterly GDP growth are negatively related to quarterly change in unemployment. It’s validity is limited to short run cycles and does not hold for longer time periods. It can more accurately be described as a heuristic as it is mostly statistical relationships and not based on theoretical prediction.
The empirical observations apply not only to % changes in unemployment but also shows responses to other labor related statistics such as hours worked and labor productivity (Hartley, 2015). The simplest explanation for the phenomenon is an analysis of the terms in Equation 1. A feature of labor markets is that hours worked per week are usually on fixed terms. Also in the short run labor productivity, population, eligible workforce are more rigid so employment hours are the most responsive (elastic) to changes in aggregate demand (GDP) (Knotek, 2007).
The theory of counter-cyclical policy is presented in Core Economics Chp 15. A casual observation of the time series of employment and GDP for any market economy will observe some periodic, up-and-down fluctuating pattern. The exact causes for these cycles is the subject of debate although a few patterns are consistent. The “paradox of thrift” (Core Economics Chp 14) is a term used by Keynes to describe the positive feedback dynamics of household consumption during a recession. In a household which has lost employment they will reduce spending and when there is uncertainty of jobs overall, this behavior spreads to other households. The contraction in spending leads to a further drop in aggregate demand, triggering more unemployment. The reason why it is called a paradox is that the thrift response from households is the right thing for them to do, but presents challenges at a systemic level for sustaining high employment for everyone. The opposite effect can also occur during credit expansion periods where more household spending leads to the creation of new jobs. This is the motivation for the state to intervene to mitigate these positive feedback cycles. The components of GDP can be split into consumption, and investments where consumption is intended for immediate use and investments are spending that is intended for some benefit in the future (OECD, 2010). Both consumption and investments fluctuate in cycles, however consumption is reported to vary more smoothly and on longer time scales than investment, which tends to be more erratic and thus can be a more significant contribution to recent downturns (OECD, 2010). In response to a sudden drop, there are a number of policy interventions that can be deployed to reverse the effects and resume full output and employment levels.
Source : (OECD, 2010 ; IMF, 2020)
In any of the policy responses the effectiveness is assessed based on the short term and long term consequences. The logic of demand management is captured in the “multiplier effect”. When “stimulus” money is injected into one part of the economy, that injection multiples into another cascading series of exchanges which at each exchange are subject to taxation, and also subject to “leakages”. Leakages can occur when the money goes out of circulation and doesn’t continue circulating or “multiplying”. This can occur in several ways, either by savings, or paying down debt that is not re-invested, buying and holding assets like commodities or real estate, imports from trade with other countries or overseas investments from mobile capital in the financial markets. The multiplier ratio is a measure of the level of increase in GDP for a $ of stimulus injected (Core Economics Chp 15). If the multiplier ratio is >1 for an extended period of time in theory the government can recover the stimulus invested back over time as increased taxation revenue to pay down the debt. The measure of success in the short term is how large the multiplier effect, how well leakages are mitigated and more importantly where the stimulus is directed. Each slowdown scenario is unique and different stimulus will have heterogeneous effects concentrated in some parts of the economy and dampened effects in other areas. A well designed response identifies the correct deficient area of the economy to target and achieves the desired stimulus recovery. The other consideration are the long term consequences - debt and inflation. Stimulus is an investment from the state which requires either drawing down reserves or borrowing from a combination of external or public debt. If there is already a high level of public debt further stimulus may compromise the creditworthiness of the government and raise borrowing costs / interest rates. There are fewer options for stimulus in scenarios of high inflation and high government debt.
The three most common measures are monetary expansion, discretionary fiscal stimulus and automatic stabilizers (OECD, 2010). A monetary expansion response is either to target a low interest rate which would stimulate borrowing and hence more money in circulation or to expand the money supply through other means. The risk of excessive monetary expansion is inflation, and this measure is limited in scenario where inflation is already high. If evidence is available that there is under-utilized “slack” in the economy such as factories that are idle, etc.. from other data sources such as the case for the recovery to the 2020 coronavirus pandemic then there may be fewer concern of short term inflation than in a “no slack” scenario (IMF, 2020). Discretionary fiscal stimulus is consistently mentioned as one of the preferred responses both by the OECD report in response to the global financial crisis (OECD, 2010) and in the recent global pandemic response by the IMF (IMF, 2020).
Fiscal stimulus is increased spending above normal levels by the government, usually at a deficit and typically financed through borrowing, although for the case of Singapore it is drawing on past reserves. According to the IMF report on the coronavirus response discretionary fiscal spending stimulus can have significantly higher multiplier effect with coefficients >1 compared to tax incentives, with fiscal consumption of goods and services higher than government investments such as infrastructure (IMF, 2020). The main limitation for fiscal stimulus is debt. While there is debate about the significance of public debt - money borrowed from the citizens, accumulated levels of external debt - money borrowed from outside the economy - can present serious risks to economic stability and can result in serious disruption shocks such as hyperinflation in the extreme scenario. Governments who already have high debt from past stimulus with high repayments and weak credit may have less room to use fiscal stimulus for future recessions.
Automatic stabilizers are rule based policies that increase spending when unemployment rises and decrease when unemployment falls. Three examples of automatic stabilizers are progressive taxes that redistribute from the wealthy to the poorer households, unemployment social safety net welfare benefits and direct rules such as an automatic spending increase as a % of GDP pegged to the unemployment rate (IMF, 2020). If social benefits and progressive taxes automatic stabilizers are used together it is possible that in an entire business cycle the budget is balanced since the progressive taxes run a surplus during the recovery period to recover the deficits from the spending increases during the recession period. Non standard methods cited in the OECD study include labor and property market interventions.
Collective wage negotiating intermediary institutions were suspected to have a dampening effect on the recovery response both to mitigate the initial shock effects, and also to slow the recovery response (OECD, 2010). A hybrid approach was reported by the Scandanavian countries which applied an early wage protection response, followed by a managed correction either to hours or wage policy and this strategy was reported to be more effective at speeding the recovery (OECD, 2010). Property market taxes were cited as another form of automatic stabilizers that increase tax revenues, recover leakages in recovery periods and could dampen future inflation (OECD, 2010).
Singapore has successfully applied fiscal stimulus in recent years (MAS, 2011 ; Lee, CNBC, 2020) but there are several limitations of counter-cyclical demand management for small open market economies to consider. The first is the limited multiplier effect for open economies with trade as a large % of GDP. Trade is a source of leakage. The reported multiplier for the Scandanavian open market economies is just over 1.0 (Schettkat, 1999). If however the policy is applied at a broader scale either globally or for a common Southeast Asia economic zone with a large level of internal trade relative to external trade the potential scale of public investments and multiplier effects could be applied at larger scale and longer periods of time. The second limitation which is less of an issue for Singapore is debt. Some level of persistent public debt may not be a concern, but continuous growth of external debt can have more serious consequences. Short periods of stimulus where investments are directed to activities that ultimately enhance human capital and productive capacity for later recovery in future periods is a financially sustainable model. If however the response of the stimulus does not lead to productive outcomes or encourages more non-essential household consumption and low savings rates then this undermines the health of the economy and the integrity of the program. Tailoring the stimulus to the problem that triggered the recession is another important condition. For example a stimulus for the coronavirus pandemic that is directed at suburban housing mortgages may not be as effective compared to a plan that is directed at restaurants.
In some crisis such as a “cost-push” where a supply chain interruption, an external shock can create unemployment in ways that cannot be remedied simply through stimulus alone such as the oil crisis of the 1970’s (Huhtamäki, 2013). The situation in the late 1970’s was a period of high inflation and high unemployment “stagflation” and in this scenario typical counter-cyclical interventions are limited because of the high inflation. The debate about the cause and remedy to the stagflation of the 1970’s called into question the basic premise of Keynesian counter-cyclical intervention and it has only recently regained credibility in its role in the recovery from recent global financial crisis. The track record of the Scandanavian countries is another source of credibility as they have continuously embraced Keynesian models for macroeconomic policy (Huhtamäki, 2013). A few lessons learned from stagflation are that fiscal policy is limited with high levels of trade and when inflation is high, that there is no one fixed Phillips curve, that ambiguous foreign exchange or interest rate policy can create uncertainty and lead to more inflation, and that accounting for structural dynamics of the economy such as energy systems may be important for more accurate modeling of shocks and designing more tailored interventions.
