Output gap measurement and the New Keynesian Phillips curve for China

The New Keynesian Phillips curve implies that the output gap, the deviation of the actual output from its natural level due to nominal rigidities, drives the dynamics of inflation relative to expected inflation and lagged inflation. This paper exploits the empirical success of the New Keynesian Phillips curve in explaining China's inflation dynamics with a new measure of the output gap. We estimate the output gap using the Bayesian multivariate Beveridge–Nelson decomposition method, based on a multivariate dynamic model featuring distinct interactions among inflation, money, and real output in China. The empirical results using quarterly data spanning 1979–2010 show that the new measure of the output gap outperforms the traditional measures in fitting the New Keynesian Phillips curve. This result provides useful insights for inflation dynamics and monetary policy analysis in China.     

Financial Innovation and Divisia Money in Taiwan: Comparative Evidence from Neural Network and Vector Error-Correction Forecasting Models

In this article a Divisia monetary index is constructed for the Taiwan economy, and its inflation forecasting potential is compared with that of its traditional simple sum counterpart. The Divisia index is adjusted in two ways to allow for the financial liberalization that Taiwan has experienced since the 1970s. The powerful artificial intelligence technique of neural networks is used and is found to beat the conventional econometric techniques in a simple inflation forecasting experiment. The preferred inflation forecasting model is achieved using networks that employ a Divisia M2 measure of money that has been adjusted to incorporate a learning mechanism to allow individuals to gradually alter their perceptions of the increased productivity of money. The explanatory power of the two innovation-adjusted Divisia aggregates dominates that of the simple sum counterpart in the majority of cases. (JEL C4 , E4 , E5 )     

The application of neural networks to the Divisia index debate: evidence from three countries

If monetary policy is to be effective in controlling the macroeconomy, accurate measurement of the money supply is essential. The conventional way of measuring the level of the money supply is to simply sum the constituent liquid liabilities of banks. However, a more sophisticated, weighted monetary index has been proposed to take account of the varying degrees of liquidity of the short-term instruments included in money. Inferences about the effects of money on economic activity may depend importantly on the choice of monetary index because simple sum aggregates cannot internalize pure substitution effects. This hypothesis is investigated in the current paper. A Divisia index measure of money is constructed for the USA, UK and Italian economies and its inflation forecasting potential is compared with that of its simple sum counterpart in each of the three countries. The powerful Artificial Intelligence technique of neural networks is used to allow a completely flexible mapping of the variables and a greater variety of functional form than is currently achievable using conventional econometric techniques. The application of neural network methodology to examine the money-inflation link is highly experimental in nature and, hence, the overriding feature of this research is one of simplicity. Superior inflation forecasting models are achieved when a Divisia M2 measure of money is used in the majority of cases. This support for Divisia is entirely consistent with findings based on standard econometric techniques reported from the respective central and Federal Reserve banks of each country. Divisia monetary aggregates appear to offer advantages over their simple sum counterparts as macroeconomic indicators. Further, the combination of Divisia measures of money with the artificial neural network offers a promising starting point for improved models of inflation.     

Monetary aggregates and core inflation: evidence from India

This study constructs a model-based core inflation for India using Divisia monetary aggregates instead of traditional money measures with the methodology proposed by Bagliano and Morana (2003) and evaluates its forecasting abilities. The core inflation derived from Divisia monetary aggregates is found to be a better leading indicator of measured inflation than the core inflation derived from traditional money measures. These results argue for a case in favour of using monetary aggregates in the construction of core inflation for policy purposes.     

The information content of the divisia monetary aggregates in forecasting inflation in the euro area

The paper investigates the performance of a set of monetary indicators, based on the Divisia money constructed for the euro area, on forecasting euro area inflation. The paper first briefly discusses on the relative information contents of the Divisia aggregates and the simple sum aggregates. The forecasting performance of the former is then examined by means of simulated out-of-sample forecasting. In addition to examining the information contents of the Divisia aggregate constructed for M3 money, the study also examines the performance of the Divisia M1 money to gain evidence on the relative performance between the broad and narrow Divisia monetary aggregates. According to the results, only some of the monetary indicators considered can significantly improve the univariate inflation forecasts. The Divisia M3 money based monetary indicators turned out to perform better than their Divisia M1 based counterparts. The result contradicts some previous evidence on the optimal level on monetary aggregation in the context of broad versus narrow money.      

通货膨胀实时预测及菲利普斯曲线的适用性

本文从实时分析的视角,基于多种退势方法的产出缺口最终估计、准最终估计和实时估计序列,分别构建了四类预测模型对我国通货膨胀率进行预测,分析了产出缺口修正效应和滞后阶数变化效应对通胀预测的影响,并进一步考察了产出缺口在通胀预测中的作用及菲利普斯曲线在通胀预测中的适用性。研究结论表明,通胀率的实时预测效果要明显比基于最终数据的差,其中滞后阶数变化效应对实时预测精度的影响大于产出缺口修正效应;尤为重要的是,尽管在最终数据的预测分析中,产出缺口的引入能够提高通胀率的预测精度,但是在实时预测中,产出缺口没有提供有价值的信息,因此"产出—通胀"型菲利普斯曲线在我国通胀实时预测中并不适用。     

The New Keynesian Phillips curve and the role of expectations: Evidence from the CESifo World Economic Survey

We provide evidence on the fit of the hybrid New Keynesian Phillips curve for selected Euro zone countries, the US and the UK. Instead of imposing rational expectations and estimating the Phillips curve by the Generalized Method of Moments, we use direct measures of inflation expectations from the CESifo World Economic Survey. Our main findings are as follows: (i) The use of survey data gives empirical results, which are more reliable than those obtained from the GMM approach. (ii) The purely forward-looking Phillips curve can be rejected in favor of the hybrid New Keynesian Phillips curve. (iii) The estimated coefficients on past inflation are higher when using survey expectations than when using the rational expectations GMM approach. (iv) It remains unclear whether real unit labor costs or a measure of the output gap should be used as a proxy for real marginal costs. (v) Theory-based restrictions lead to an improvement of the empirical results.     

Banking and the role of money in the business cycle

This paper enriches a standard New Keynesian model with a simple banking sector to investigate the role of money in the business cycle. Maximum likelihood estimation of the model suggests that money balances play a significant role in explaining the intertemporal allocation of consumption and the dynamics of inflation as described by the forward-looking IS and Phillips curves. Nonetheless, the responses of the model’s variables to shocks remain qualitatively similar to a model without money, suggesting that the omission of money balances leaves the model’s transmission mechanism unaffected.     

Does a factor Phillips curve help? An evaluation of the predictive power for U.S. inflation

This article evaluates various models’ predictive power for U.S. inflation rate using a simulated out-of-sample forecasting framework. The starting point is the traditional unemployment Phillips curve. We show that a factor Phillips curve model is superior to the traditional Phillips curve, and its performance is comparable to other factor models. We find that a factor AR model is superior to the factor Phillips curve model, and is the best bivariate or factor model at longer horizons. Finally, we investigate a New Keynesian Phillips curve model, and find that its forecasting performance dominates all other models at the longer horizons.     

Do search frictions matter for inflation dynamics?

We assess the empirical relevance for inflation dynamics of accounting for the presence of search frictions in the labor market. The new Keynesian Phillips curve explains inflation as being mainly driven by current and expected future marginal costs. Recent empirical research has emphasized different measures of real marginal costs to be consistent with observed inflation persistence. We argue that, allowing for search frictions in the labor market, real marginal cost should also incorporate the cost of generating and maintaining long-term employment relationships, along with conventional measures, such as real unit labor costs. In order to construct a synthetic measure of real marginal costs, we use newly available labor market data on worker finding and separation rates that reflect hiring and firing costs. We then estimate a new Keynesian Phillips curve by generalized method of moments (GMM) using the imputed marginal cost series as an observable and find that the contribution of labor market frictions in explaining inflation dynamics is small.     

PRODUCTIVITY GROWTH AND THE PHILLIPS CURVE: A REASSESSMENT OF THE US EXPERIENCE

In this paper, we analyse a new Phillips curve (NPC) model and demonstrate that (i) frictional growth, i.e., the interplay of wage staggering and money growth, generates a non‐vertical NPC in the long run and (ii) the Phillips curve (PC) shifts with productivity growth. On this basis we estimate a dynamic system of macro‐labour equations to evaluate the slope of the PC and explain the evolution of inflation and unemployment in the USA from 1970 to 2006. Because our empirical methodology relies heavily on impulse response functions, it represents a synthesis of the traditional structural modelling and (structural) vector autoregressions. We find that the PC is downward‐sloping with a slope of ?3.58 in the long run. Furthermore, during the stagflating 1970s, the productivity slowdown contributed substantially to the increases in both unemployment and inflation, whereas the monetary expansion was quite ineffective and led mainly to higher inflation. Finally, the monetary expansion and productivity speedup of the roaring 1990s were both responsible for the significant lowering of the unemployment rate.     

公众预期、需求冲击与通货膨胀——兼论通胀持久性的地区差异

本文基于具有微观基础的混合式新凯恩斯菲利普斯曲线模型,利用1985-2009年的省际面板数据,分析了预期因素、需求冲击及产出缺口与通货膨胀之间的动态过程及其地区差异。研究发现：总体上,无论是适应性预期还是理性预期对当期通货膨胀都有较强的推动作用;作为需求冲击的居民消费支出、固定资本形成以及产出缺口对通货膨胀都具有正向拉动作用。从通货膨胀持久性的地区差异上看,西部地区的通胀持续性最长,中部最短。     

Does money have a role in the inflation process? Evidence from Australia

Inflation, defined as a sustained increase in the price level, is considered a monetary phenomenon, as it can be explained within the framework of money‐demand and money‐supply relationships. In the extant literature, money growth is shown to remain causally related to inflation across countries and over time, irrespective of the exchange rate regime and stability of the money‐demand function. Nevertheless, emerging literature suggests a diminishing role of money in the conduct of monetary policy for price stability, especially under inflation targeting. Monetary policy in Australia under inflation targeting since 1993 is an example of policy that denies a relationship between money growth and inflation. The proposition that money does not matter insofar as inflation is concerned seems odd in both theory and the best‐practice monetary policy for price stability. This paper uses annual data for the period 1970–2017 and quarterly data for the period 1970Q1–2015Q1. It deploys both the Johansen cointegration approach and the autoregressive distributed lag (ARDL) cointegration approach to investigate for Australia whether money, real output, prices and the exchange rate (non‐stationary variables) maintain the long‐run price‐level relationship that the classical monetary theory suggests in the presence of such stationary variables as the domestic and foreign interest rates. As expected, the empirical findings for Australia are consistent with the classical long‐run price‐level relationship between money, real output, prices and the exchange rate. The error‐correction model of inflation confirms the presence of a cointegral relationship among these variables; it also provides strong evidence of a short‐run causal relationship between money supply growth and inflation. On the basis of a priori theoretical predictions and empirical findings, the paper draws the conclusion that the monetary aggregate and its growth rate matter insofar as inflation is concerned, irrespective of the strategy of monetary policy for price stability.     

Response to Seitz and Tödter, 'How the P* Model Rationalizes Monetary Targeting: A Comment on Svensson'

Seitz and TÎdter argue, counter to Svensson, that the P * model provides a rationale for money-growth targeting. In particular, they argue that 'money growth targeting is a special form of inflation forecast targeting based on a "limited" information set. In contrast to "full information" inflation forecast targeting, money growth targeting is likely to be more robust under changing conditions of the real world'.
However, money-growth targeting is better described as a special case of inflation targeting, namely when money growth is considered to be the only predictor of future inflation. But there is overwhelming empirical evidence that there are not only other, but also better, predictors of future inflation than money growth, which makes inflation-forecast targeting superior to money-growth targeting. Inflation-forecast targeting is indeed more robust (in the sense of using available information and allowing judgemental adjustments in a flexible way) than monetary targeting.
In particular, in the P * model, the real money gap is a better predictor of future inflation than money growth, as demonstrated theoretically by Svensson and empirically by Gerlach and Svensson (the empirical finding is also confirmed by Trecroci and Vega). Therefore, inflation-forecast targeting is superior also within the P * model. Under `changing conditions of the real world', for instance after the formation of a monetary union, money growth is likely to be particularly unreliable as a predictor of future inflation, making monetary targeting especially unsuitable and non-robust.     

A Search Model of Unemployment and Inflation*

This paper introduces money into the standard labor‐matching model. A double‐coincidence problem makes money necessary as a medium of exchange. In the long run, a rise in the growth rate of money leads to higher inflation and higher unemployment, such that the long‐run Phillips curve is not vertical. The optimal monetary growth rate decreases with greater worker bargaining power, the level of unemployment benefits, and the payroll tax rate.     

Why Has Australian Wages Growth Been So Low? A Phillips Curve Perspective

Wages growth in Australia has recently been the lowest in two decades. One possible explanation is a decline in the non‐accelerating inflation rate of unemployment (NAIRU). We examine this hypothesis by estimating a wage Phillips curve including a time‐varying NAIRU. Our findings are: (i) the NAIRU has recently been around 5.5 per cent; (ii) our approach increases the precision of the NAIRU estimates; (iii) low inflation expectations have been an important contributing factor; and (iv) the long‐run annual wages growth is nearly 3 per cent. We also find that the underutilisation rate suggests greater slack exists, but is less useful in explaining wage developments.     

Price Expectations and the Australian Price Level: 1901–30*

This paper examines the sources of change in Australian prices in the first three decades of this century. It probes the information content of simple ‘money is dominant’ models of inflation and of the expectations-augmented Phillips curve and shows that price expectations played a dramatically different role then than in the 1970s.     

The stability of the Turkish Phillips curve and alternative regime shifting models

This article, presents empirical evidence supporting instability of the Phillips curve in Turkey. We employ the multiple structural break models and the Markov-switching models and then evaluate the performance of the two models. The data pertain to the monthly inflation rate in Turkey for the period of 1987 to 2004. The results show that the Turkish Phillips curve is not linear. There exists no evidence on the asymmetry in the inflation response to output gap. The persistence of inflation is found to be much lower than in linear models. After 2001, slight decline in persistence of inflation is observed. There exits weaker support for the Phillips curve for the periods where the policymakers attempt to take the advantage of the tradeoff between output and inflation relationship.     

Inflation Dynamics in the Euro Area and the Role of Expectations

This paper examines the empirical performance of the New Keynesian Phillips curve and its hybrid specification in the euro area. Instead of imposing rational expectations, direct measures, i.e. OECD forecasts, are used as empirical proxies for economic agents’ inflation expectations. Real marginal costs are proxied by three alternative measures. The results suggest that once the rational expectations hypothesis is relaxed and directly measured expectations are used, the European inflation process can be modeled using the forward-looking New Keynesian Phillips curve. However, when allowing for possible non-rationalities in expectations, inflation can be modeled more accurately by the hybrid Phillips curve with the additional lagged inflation term. In this approach, output gap turns out to be at least as good as labor income share as a proxy for real marginal cost. Moreover, the inflation process seems to have become more forward-looking in the recent years of low and stable inflation.The views expressed are those of the author and do not necessarily reflect the views of the Bank of Finland. Special thanks are due to the editor, two anonymous referees, Juha Tarkka, Jouko Vilmunen and Matti Virén for useful comments. I am also grateful to David Mayes and Geoffrey Wood for helpful suggestions and to Heli Tikkunen for excellent research assistance. For their constructive comments, I would also thank participants in the conference on the Eurosystem Inflation Persistence Network at the ECB, which was held in Frankfurt in December 2003.     

STRUCTURAL CHANGE AND NONLINEARITIES IN A PHILLIPS CURVE MODEL FOR SOUTH AFRICA

The main objective of this article is to reexamine the role of the Phillips curve for monetary policy analysis in South Africa by augmenting the model for major structural changes in the balance-of-payments and labor market. The main findings show that a linear Phillips curve with an output gap in levels accurately describes South Africa's nontrended inflation experience during 1971(Q1)–1984(Q4), whereas a piecewise concave curve with an output gap in growth rates correctly predicts the decelerating inflation pattern during 1986(Q1)–2001(Q2). The concave curve after 1985 imparts a deflationary bias that requires expansionary demand-side policies to stabilise the inflation rate. An important corollary is that expansionary demand-side policies can raise the average growth rate of the output gap over time without sacrificing stabilization objectives. (JEL C22, E3, E52 )