Monetary policy, as currently being implemented, is failing to deal with the current raft of economic issues, including low inflation, stagnant wage growth, high asset prices and ultra low policy rates. When coupled with politicians taking a back seat and their inability to tackle the core issues, macroprudential measures are being tried, in a massive real-time experiment. This coupling of monetary and macroprudential action is largely untested. Can they work in tandem?
A Bank of England working paper “Monetary versus macroprudential policies causal impacts of interest rates and credit controls in the
era of the UK Radcliffe Report“, attempts to look at this issues, with some interesting results. They conclude that macroprudential policy is better suited to achieving financial stability goals than monetary policy.
The Global Financial Crisis and its disappointing aftermath are widely viewed as a major macroeconomic policy disaster from which lessons must be learned. Yet agreement on the precise failures and, thus, the necessary lessons, has been elusive in many areas, from mortgage regulation to fiscal policy, and from global imbalances to central banking. In the latter case, the role of macroprudential policies remains fraught, with doubts about whether they should exist, if they work, and how they should be designed and used.
Reflecting this range of skepticism, several countries have recently taken quite varied courses of action in retooling their policy regimes since 2008. For example, facing a heating up of their housing markets in 2010–12, Sweden and Norway took quite different policy actions. Sweden’s Riksbank tried to battle this development using monetary policy tools only, raising the policy rate, and tipping the economy into deflation, as had been predicted by the dissident Deputy Governor Lars Svensson, who subsequently resigned. Across the border, the Norges Bank implemented some cyclical macroprudential policies to crimp credit expansion and moderate mortgage and house-price booms, without relying as much on rate rises, and they managed to avoid such an out-turn. Elsewhere, other countries display differing degrees of readiness or willingness to use time-varying macroprudential policies. The Bank of England now has both a Financial Policy Committee and a Monetary Policy Committee, and the former has already taken macroprudential policy actions under Governor Mark Carney. As Governor of the Bank of Israel, Stanley Fischer utilized macroprudential policies against perceived housing and credit boom risks, but now as Vice-Chair at the Federal Reserve his speeches lament the lack of similarly strong and unified macroprudential powers at the U.S. central bank. Yet as one surveys these and other tacks taken by national and international policymakers, two features of the post-crisis reaction stand out: the extent to which these policy choices have proved contentious even given their limited scope and span of operation, and the way that the debate on this policy revolution has remained largely disconnected from any empirical evidence. And of course, the two features may be linked.
This paper seeks a scientific approach that might address both of these shortcomings, by bringing a new and vastly larger array of formal empirical evidence to the table. To that end, we turn to the last great era of central bank experimentation with the same types of macroprudential instruments: the postwar decades from the 1950s to the early 1980s when many types of credit controls were put in play. We go back and construct by hand new quantitative indicators on the application of such policies in the UK, including credit ceilings, hire purchase regulations, special deposits, and the “Corset.” To evaluate the impacts of these policies, and to compare them with the impacts of the standard monetary policy tool of Bank Rate, we then implement a state-of-art econometric estimation of impulse-response functions (IRFs) to the two policy shocks by developing a new approach to identification that is also original to this paper, one that we shall refer to as Factor-Augmented Local Projection (FALP). Our approach unites the flexible and parsimonious local projection (LP) method of estimating IRFs with the Romer and Romer approach of using forecasts to mitigate the selection bias arising from policymakers acting on their expectations of future macroeconomic developments. To ensure greater robustness, we also borrow from the factor augmentation approach that has been employed in the VAR literature as a means to control for other information correlated both with changes in policy and future macroeconomic developments. We subject our results to a range of robustness tests, most of which give us little reason to doubt our main results.
We report three main results. First, we find that monetary and credit policies had qualitatively distinct effects on headline macroeconomic indicators during this period. Increases in Bank Rate had robust negative effects on manufacturing output, and consumer prices especially, and positive effects on the trade balance. However, the estimated response of bank lending to an increase in Bank Rate is not statistically significant. By contrast, we find that credit controls — liquidity requirements on banks, credit growth limits, and constraints on the terms of consumer finance — had a strong negative impact on bank lending. We also find some evidence that credit policies may have depressed output, improved the trade balance, and led to an increase in consumer prices. But our evidence here is less strong than for corresponding shifts in Bank Rate. Overall, our estimates suggest that monetary and credit policies spanned different outcome spaces during this period. This result supports the notion that today’s macroprudential tools, which are close cousins of the credit policies studied in this paper, might provide the additional independent tools required to help central banks meet both their monetary and financial stability objectives.
Second, we find that our estimated monetary and credit policy shocks were major drivers of macroeconomic dynamics over the 1960s and 1970s. A significant fraction of lending and output dynamics can be explained by these shocks. Moreover, we find that a large fraction of the pick-up in inflation in the 1970s can be attributed to expansionary monetary policy shocks — that is, interest rates were substantially looser in the latter part of our sample than would have been expected given available econonomic forecasts and the information about the state of the economy contained in our estimated factors.
Third, our impulse responses indicate that credit policies had moderating effects on modern-day indicators of financial system vulnerabilities, while the effects of monetary policy actions were less clear cut. Contractionary credit policies had large and persistent negative effects on the credit-to-GDP ratio; they also reduced banks’ loan to deposit ratios (a measure of their resilience), and increased the spread between debentures i.e. term corporate debt instruments and gilts (a measure of investor risk appetite). In contrast, we find that contractionary monetary policy led to a persistent increase in the credit-to-GDP ratio, as the fall in GDP exceeded the fall in credit. Contractionary monetary policy actions also led to a small reduction in banks’ loan-to-deposit ratios, but led to a large persistent increase in the debenture spread. Our results therefore provide some support to the view that macroprudential policy is better suited to achieving financial stability goals than monetary policy.
Note: Staff Working Papers describe research in progress by the author(s) and are published to elicit comments and to further debate. Any views expressed are solely those of the author(s) and so cannot be taken to represent those of the Bank of England or to state Bank of England policy. This paper should therefore not be reported as representing the views of the Bank of England or members of the Monetary Policy Committee, Financial Policy Committee or Prudential Regulation Authority Board
