Rethinking monetary policy: The case for nominal GDP targeting

The monetary policy environment has changed considerably since the mid-twentieth century, from the development of economic thinking to the changes in policy priorities and the lessons learned from past experiences. Each transition, from the demise of the gold standard to floating exchange rates, monetary targeting and, more recently, the introduction of inflation targeting, has been a reactive response to the prevailing economic conditions of the time, whether driven by inflation, financial stability concerns, or geopolitical pressures. Despite these changes, the question must be asked: is there a better framework to ensure long-term price stability and economic growth? Since the early 1990s inflation targeting has been at the heart of contemporary monetary policy. This regime aims to maintain price stability by keeping inflation close to a medium-term target, generally around 2 per cent annually. While widely adopted by central banks of advanced economies as the global standard, the limitations of this framework have been exposed by financial crises and supply-side shocks. Too often, central banks have underestimated the influence of fiscal policy and their own balance sheet expansions on inflation trends, contributing to monetary policy decisions that have added to the erosion of real incomes and exacerbated the cost-of-living crisis. I believe we should have a fundamental rethink of the current monetary policy framework. Central banks should implement a framework that targets the growth rate or level path of nominal GDP (NGDP)1 rather than relying on an inflation target of 2 per cent. NGDP targeting (NGDPT) offers a more transparent, rule-based approach that reduces subjectivity and the risk of policy errors by looking to stabilise total nominal spending in the economy. In contrast, the current ‘constrained discretion’2 framework gives policymakers considerable flexibility (discretion) to react to fluctuating economic conditions within loosely defined parameters (constraints). By reducing reliance on discretionary decision-making, NGDPT would increase predictability and transparency for financial markets, thereby allowing central banks to regain any lost credibility. NGDPT, also known as nominal income targeting, provides a more flexible and adaptive framework for managing external and supply-side shocks, such as the Covid-19 pandemic and energy price spikes, which contributed to driving inflation to its highest levels in four decades.3 I argue here that NGDPT is better equipped to maintain long-term economic stability by reducing fluctuations in output and employment more effectively. By reducing reliance on discretionary decision-making, NGDPT would provide a free-market approach to monetary policy and give households and firms clearer guidance on the future path of interest rates. My objective is to demonstrate how an NGDPT-based framework could address the deficiencies of the current system. A key flaw of inflation targeting is its inability to distinguish between demand- and supply-driven inflation, leading to suboptimal policy responses during economic shocks. By advocating NGDPT, this article seeks to add to the broader debate on the future direction of monetary policy. Inflation targeting was first introduced in New Zealand in 1990 (see Reserve Bank of New Zealand Act 1989), followed by Canada in February 1991, and the Bank of England in October 1992. This framework rests on two key pillars: (a) a flexible approach for discretionary responses to short-term economic shocks, and (b) a clear numerical objective for inflation to anchor expectations and ensure price stability. Although there are many different definitions of inflation targeting, the fundamental idea is that a forward-looking central bank is committed to maintaining inflation at a predetermined target, usually around 2 per cent, by conducting monetary policy through the interest rate channel.4 Changes to a short-term interest rate are the central bank's primary instrument, affecting aggregate demand to bring actual inflation into line with the inflation target. The underlying assumptions are that markets are flexible, that prices and wages adjust quickly to changes in monetary conditions, and that households and firms form rational expectations for both short-term and future inflation.5 By committing to a clear 2 per cent inflation target, policymakers provide a secure anchor for inflation expectations by lowering uncertainty regarding future price levels, at least over the medium term (see Gürkaynak et al., 2006; 2007; Ravenna, 2007). Firms are more confident when making medium- to long-term investment decisions thanks to this stability, and consumers are able to plan their spending and saving decisions more effectively. Orthodox monetary policy suggests that central banks should initially ‘look through’ short-term inflation caused by adverse supply shocks, given their temporary impact on output.6 If central banks react too aggressively to such shocks, they risk amplifying economic volatility rather than stabilising it. Under ‘constrained discretion’, central banks can temporarily deviate from their rigid inflation target to support broader economic objectives such as employment and financial stability, although to what extent is subjective. Despite some successes, inflation targeting has attracted a good deal of criticism for its rigidity and narrow focus, which often neglects other important macroeconomic variables, not to mention wider financial stability concerns. This limited scope may result in suboptimal policy decisions, especially during financial crises. Inflation targeting also ignores asset prices, particularly those of property and equities, which can result in asset bubbles, adding to the frequency of boom–bust cycles. The debate over the best (and most agile) monetary system is continuing against a backdrop of an unstable macroeconomic climate, which includes very high levels of private and national debt, chronically low productivity growth in some economies, de-globalisation, and an ageing population with associated social care costs. Furthermore, we are currently experiencing a period of increased geopolitical tensions. One measure that has gained popularity in recent years is NGDPT. By focusing on real output and inflation, NGDPT aims to stabilise overall economic growth rather than the rate of inflation. A fundamental principle in the creation of monetary policy is the requirement for a quantitative anchor: a baseline that guides market expectations and policy actions. According to Fatás et al. (2007), stability and predictability in monetary policy depend on a reliable anchor, be it inflation, the money supply, or another measure. According to Frankel (2012), NGDPT offers a more balanced approach to managing economic volatility, since it naturally adapts to both supply- and demand-side shocks. Under this approach, monetary policy will permit some deflationary pressure if output increases and productivity rises, but will tolerate higher-than-target inflation to support recovery during a downturn when growth is below the desired target. Although some economists and central bankers find this framework to be an appealing alternative, this has not yet become the consensus view, despite growing attention in policy and media circles. Furthermore, to date, it has not been adopted by any central bank. Its slow acceptance in scholarly discussions reflects concerns about practical applicability, especially with respect to data accuracy, policy transmission, and public understanding. However, with recent crises exposing shortcomings in inflation targeting, and with many economies flirting with stagnation and therefore a renewed focus on growth, NGDPT should be re-evaluated as a viable alternative to inflation targeting. NGDP targeting traces its intellectual origins to early proponents such as Meade (1978), Tobin (1980), Bean (1983), and Gordon (1985) – although others argue its roots lie in the work of Hayek and Austrian economics (see Evans, 2016). It emerged as a compelling alternative to the post-war Keynesian consensus, with its strong focus on demand management, and to the monetary targeting policies of the 1970s, particularly due to its ability to counteract velocity shocks (i.e. shifts in money demand) (see Beckworth, 2019). The debate between rule-based and discretionary approaches to monetary policy has regained attention recently, particularly after the criticisms of policy responses following the Covid-19 pandemic. Rules (intermediate targets) act as nominal anchors for central banks, committing them to implement consistent policies. Advocates of a rule-based approach, such as Kydland and Prescott (1977), argued that although discretionary policies may address immediate issues such as unemployment, they often result in ‘inflationary bias’ and greater long-term economic instability. Discretionary inflation bias, as described by Kydland and Prescott (1977) and Barro and Gordon (1983), arises when central banks attempt to stimulate employment above its natural level by surprising markets with inflation. However, this approach typically leads to higher inflation without reducing unemployment, and weakens central bank credibility. When policymakers deviate from their commitments to react to short-term conditions, they risk undermining trust, which can lead to elevated inflation expectations among market participants and greater inefficiencies in wage and price setting. Such time-inconsistency problems – where policymakers deviate from pre-committed policies for short-term gains – affect both monetary and fiscal policy decisions. When investors anticipate policy deviations, they demand higher yields on government debt, increasing borrowing costs and contributing to fiscal inefficiencies. Inconsistent policies can also lead to ‘stabilisation bias’, resulting in greater volatility in inflation and economic output, increasing uncertainty and complicating long-term business planning. Therefore, rules are preferable to discretion in monetary policy. Proponents of NGDPT argue that its rule-based approach enhances predictability and transparency, while providing a countercyclical mechanism to counter recessions and booms, and reducing the risk of destabilising inflation bias. This systematic approach minimises uncertainty and allows market participants better to anticipate central bank actions. By contrast, the current system of constrained discretion has been criticised for its lack of clarity. The delayed response of central banks to rising inflation in 2021, initially downplayed as ‘transitory’, demonstrates how discretionary policies can lead to suboptimal outcomes. NGDP targeting can be implemented by targeting either changes in growth rates or a price level. McCallum (2011) argues that targeting a growth rate without commitment to returning to the target path for deviations (‘let bygones be bygones’) aligns more with a rule-based approach consistent with inflation targeting. By contrast, level targeting involves returning NGDP to its pre-deviation path (bygones are not bygones). Level targeting would solve the problem of ‘base drift’7 to which a focus on growth rates is susceptible, although it would also require more frequent and discretionary adjustments. In the UK, the Bank of England already incorporates growth rate targeting within its inflation targeting framework. Shifting to NGDP level targeting would therefore likely involve substantial changes, requiring some experimentation and potentially more discretionary policy interventions. Level targeting also presents communication challenges, as the public may struggle to understand the ‘make-up’ actions required to account for past deviations from its target path, complicating the Bank's messaging. However, some economists argue that level targeting could improve policy effectiveness by providing greater certainty about the future trajectory of NGDP. By encouraging financial markets to anticipate corrective make-up measures to return NGDP to its target path, level targeting could reduce economic volatility through clearer forward guidance from the central bank. Level targeting is not new. In some ways, the US Federal Reserve's actions resemble price level targeting within its existing ‘flexible average inflation targeting’ (FAIT) monetary policy framework.8 Under this framework, the Federal Reserve is committed to seek to achieve inflation that averages 2 per cent over time (Powell, 2020), adjusting for any deviations above or below this target. For example, if inflation in the first year is zero and 4 per cent in the second year, inflation has averaged 2 per cent over the two years. This would be consistent with a price-level target that accommodates periods of above- and below-target inflation with contractionary and expansionary monetary policies to stay on the desired target path. A growth rate strategy aims to stabilise the growth of total nominal output – combining the desired rate of inflation and growth rate of potential real GDP9 – by setting a target nominal growth rate. Table 1 illustrates this. If potential output growth (Y) in the economy is 3 per cent, and the desired level of inflation (P) is 2 per cent, policymakers would set the nominal growth target (PY) at 5 per cent. If potential output were to fall to 1 per cent, inflation would need to rise to 4 per cent to maintain the NGDP target until output normalises. The dual focus of NGDPT reduces the volatility of the output gap (the difference between actual and potential output) and smooths fluctuations in the business cycle. By targeting steady growth for NGDP, central banks can stabilise employment more effectively and support economic activity, even during crises. In the event of an adverse supply shock (e.g. a rise in prices but reduced output below potential, including an adverse effect on the labour market), NGDPT would allow inflation to rise temporarily without triggering restrictive monetary policy that could exacerbate a downturn. During a demand-driven downturn, the same target would encourage more aggressive monetary easing to support growth and employment. The goal is to have less overall volatility in the economy, even if it results in greater volatility in short-term inflation. While inflation targeting works well in dealing with demand shocks, it is ineffective when faced with supply-side shocks, in which output and inflation move against each other. This difficulty was demonstrated during the Covid-19 pandemic, when supply chain interruptions added to increased costs while reducing output. Economies were further harmed by the decision of central banks to tighten monetary policy to fight inflation during this period because nominal wages could not keep up with rising prices. NGDPT offers a more robust framework to handle supply shocks. By focusing on stabilising NGDP, policymakers can allow the price level to adjust to changes in output without immediately tightening policy. Using our previous example and a growth target of 5 per cent for NGDP, if an adverse supply shock caused the growth rate to fall below 5 per cent, a central bank would make monetary policy more accommodative by lowering interest rates. However, under inflation targeting the movement in interest rates would depend solely on what was happening to inflation. For example, if inflation had increased to 3 per cent (i.e. above its 2 per cent target), policymakers would raise interest rates regardless of growth falling below its desired target. A central bank following a level target would have to follow an accommodative monetary policy until such time that NGDP returned to its initial 5 per cent growth level path. This would require NGDP growth to exceed 5 per cent for a period. NGDPT offers a more effective alternative to the serious problem conventional monetary policy encounters when faced by the zero lower bound to interest rates, by focusing on stabilising NGDP, which implicitly manages the money supply and its velocity.10 This approach functions as a velocity-adjusted money supply target, where changes in the money supply balance changes in money demand (See Tobin, 1983; Woodford, 2012). When economic uncertainty increases money hoarding, NGDPT would automatically loosen monetary policy, encouraging spending. Conversely, when spending accelerates, it would tighten conditions. This flexibility preserves stability more successfully than quantitative easing alone. If NGDP falls below target, policymakers could implement a prolonged monetary stimulus, allowing inflation to rise temporarily and incentivising firms to increase production (being profit seeking), ultimately driving real GDP above potential. A proposal by Scott Sumner (1989; 1995), a noted advocate for NGDPT, involves establishing a subsidised NGDP futures market. Policymakers would set a target growth rate for NGDP, say 5 per cent annually, and allow market participants to trade NGDP futures contracts based on their expectations of future NGDP performance. Traders would purchase futures in anticipation of higher-than-target growth if they believed that NGDP would surpass the target. On the other hand, they would sell futures if they anticipated that growth would be insufficient. To maintain market stability, the central bank would be prepared to purchase or sell an unlimited number of NGDP futures contracts at the target price to those who wanted to take a short or long position. Policymakers would conduct an open-market operation (OMO) in the opposite direction for each purchase or sale of these contracts. This would ensure that the money supply reflected market expectations of NGDP growth. For instance, if the central bank bought futures contracts, it would buy assets to boost liquidity in the economy; if it sold futures, it would sell assets to reduce the money supply. Market expectations would drive this strategy, which would continuously modify monetary policy to achieve the NGDP target. This would reduce the need for central bank discretionary actions and provide policymakers with real-time feedback on economic performance. To reduce central bank counterparty risk, Sumner suggests mandating that participants deposit ten cents per futures contract (covering an approximate 10 per cent change in NGDP) in a margin account. In this way, excessive risk taking would be avoided and the central bank would be protected from significant financial exposure. Despite its theoretical appeal, NGDPT presents a number of practical problems with implementation. One of the main issues is the challenge in obtaining accurate, real-time GDP data; as Orphanides and van Norden (2002) note, GDP data are typically available only on a quarterly basis and are frequently revised, which means that initial estimates may be inaccurate and could result in poor policy decisions. As a result, questions have been raised about the practicality of an NGDPT framework for monetary policy. Scott Sumner (2013) addresses these concerns with the creation of an NGDP futures market that would target the forecast, as outlined above. Another criticism is the possibility of increased inflation volatility. Stabilising NGDP could make inflation more unpredictable, particularly if real output fluctuates significantly. Central banks may be concerned that changes in inflation make it more difficult to maintain price stability in the absence of a stable inflation anchor. Adopting NGDPT would also require a significant improvement in communication by central banks, as households and firms do not usually scrutinise nominal GDP data. Frankel (2012) has suggested a phased transition, beginning with NGDP target ranges alongside inflation targets, allowing time for markets to adjust to the new approach while minimising disruption. No single monetary policy framework can comprehensively manage the complexities of a contemporary economic system. Inflation targeting, as in previous regimes, should not be seen as permanent. Its sole focus on one macroeconomic variable is a design flaw that no longer makes it the optimal anchor for monetary policy. I have tried in this article to demonstrate the effectiveness of NGDPT as a compelling alternative to inflation targeting for conducting monetary how it can better both supply-side and demand-side shocks, how it automatically incorporates economic growth into monetary policy decisions, and its ability to deal with the zero lower of these make it a serious alternative for the monetary policy framework and long-term financial and economic stability. by providing a clear and approach to monetary policy while central bank NGDPT would increase central bank credibility. However, such a of practical challenges, particularly to communication and data Central banks should move a monetary system that reduces on discretionary NGDP targeting a in that but the goal should be a framework where market not the supply of Reserve Bank of New Zealand Act 5