Working papers
(with Fernando Eguren-Martin, Guido Maia and Sevim Kösem)
We propose a novel approach to extract factors from large data sets that maximize covariation with the quantiles of a target distribution of interest. From the data underlying the Chicago Fed's National Financial Conditions Index, we build targeted financial conditions indices for the quantiles of future US GDP growth. We show that our indices yield considerably better out-of-sample density forecasts than competing models, as well as insights on the importance of individual financial series for different quantiles. Notably, leverage indicators appear to co-move more with the median of the predictive distribution, while credit and risk indicators are more informative about downside risks.
(with Michael Kumhof, Marco Pinchetti and Phurichai Rungcharoenkitkul)
BIS Working Paper 1086, CEPR Discussion Paper 17982 (Under review) [SUERF column] [VoxEU column] [Bloomberg Terminal piece]
We study the consequences for business cycles and welfare of introducing an interest-bearing retail CBDC, competing with bank deposits as medium of exchange, into an estimated 2-country DSGE environment. CBDC issuance of 30% of GDP increases output and welfare by around 6% and 2%, respectively. Financial shocks account for around half of the variance of aggregate demand and inflation, and for the bulk of the variance of financial variables. An aggressive Taylor rule for the interest rate on reserves achieves welfare gains of 0.57% of steady state consumption, an optimized CBDC interest rate rule that responds to a credit gap achieves additional welfare gains of 0.44%, and further gains of 0.57% if accompanied by automatic fiscal stabilizers. A CBDC quantity rule, a response to an inflation gap, a cash-like CBDC, and CBDC as generalized access to reserves, yield significantly smaller gains. CBDC policies can substantially reduce the volatilities of domestic and cross- border banking flows and of the exchange rate. Optimal policy requires a steady state quantity of CBDC of over 40% of annual GDP.
ECB Working Paper 2624 (Under review)
I propose a new model, conditional quantile regression (CQR), that generates density forecasts consistent with a specific view of the future evolution of some variables. This addresses a shortcoming of existing quantile regression-based models, for example the at-risk framework popularised by Adrian et al. (2019), when used in settings, such as most forecasting processes within central banks and similar institutions, that require forecasts to be conditional on a set of technical assumptions. Through an application to house price inflation in the euro area, I show that CQR provides a viable alternative to existing approaches to conditional density forecasting, notably Bayesian VARs, with considerable advantages in terms of flexibility and additional insights that do not come at the cost of forecasting performance.
(with Zymantas Budrys and Mario Porqueddu)
ECB Working Paper 2602 (Under review) [Slides] [VoxEU column] [ECB Research Bulletin]
We quantify the effects of wage bargaining shocks on macroeconomic aggregates using a structural vector auto-regression model for Germany. We identify exogenous variation in bargaining power from episodes of minimum wage introduction and industrial disputes. This narrative information disciplines the impulse responses to a wage bargaining shock of unemployment and output, and sharpens inference on the behaviour of other variables. The implied transmission mechanism is in line with the theoretical predictions of a large class of search and matching models. We also find that wage bargaining shocks explain a sizeable share of aggregate fluctuations in unemployment and inflation, that their pass-through to prices is very close to being full, and that they imply plausible dynamics for the vacancy rate, firms' profits, and the labour share.
(with Fernando Eguren-Martin, Cian O'Neill and Lukas Von dem Berge)
ECB Working Paper 2538, BOE Working Paper 881 (Under review) [Blog post]
We characterise the probability distribution of capital flows for a panel of emerging market economies conditional on information contained in financial asset prices, with a focus on 'tail' events. Our framework, based on the quantile regression methodology, allows for a separate role of push and pull-type factors, and offers insights into the term-structure of these effects. We find that both push and pull factors have heterogeneous effects across the distribution of capital flows, with the strongest reactions in the left tail. Also, the effect of changes in pull factors is more persistent than that of push factors. Finally, we explore the role of policy, and find that macroprudential and capital flow management measures are associated with changes in the distribution of capital flows.
(with Jakub Chalmoviansky and Mario Porqueddu)
ECB Working Paper 2501
We compare direct forecasts of HICP and HICP excluding energy and food in the euro area and five member countries to aggregated forecasts of their main components from large Bayesian VARs with a shared set of predictors. We focus on conditional point and density forecasts, in line with forecasting practices at many policy institutions. Our main findings are that point forecasts perform similarly using both approaches, whereas directly forecasting aggregate indices tends to yield better density forecasts. In the aftermath of the Great Financial Crisis, relative forecasting performance was typically only affected temporarily. Inflation forecasts made by Eurosystem/ECB staff perform similarly or slightly better than those from our models for the euro area.
(with Michael Kumhof and Phurichai Rungcharoenkitkul)
BOE Working Paper 884, BIS Working Paper 890, CEPR Working Paper 15526 (Under review) [Slides] [Blog post] [The Economist] [World Economics Roundtable] [Discussion by Maurice Obstfeld]
Understanding gross capital flows is increasingly viewed as crucial for both macroeconomic and financial stability policies, but theory is lagging behind many key policy debates. We fill this gap by developing a two-country DSGE model that tracks domestic and cross-border gross positions between banks and households, with explicit settlement of all transactions through banks. We formalise the conceptual distinction between cross-border saving and financing, which often move in opposite directions in response to shocks. This matters for at least four policy debates. First, current accounts are poor indicators of financial vulnerability, because in a crisis, creditors stop financing debt rather than current accounts, and because following a crisis, current accounts are not the primary channel through which balance sheets adjust. Second, we reinterpret the global saving glut hypothesis by arguing that US households do not finance current account deficits with foreigners’ physical saving, but with digital purchasing power, created by banks that are more likely to be domestic than foreign. Third, Triffin’s current account dilemma is not in fact a dilemma, because the creation of additional US dollars requires dollar credit creation by US and non-US banks rather than US current account deficits. Finally, we demonstrate that the observed high correlation of gross capital inflows and outflows is overwhelmingly an automatic consequence of double entry bookkeeping, rather than the result of two separate sets of economic decisions.
(with Ambrogio Cesa-Bianchi, Michael Kumhof and Gregory Thwaites)
BOE Working Paper 817 [Slides]
We study exchange rate determination in a 2-country model where domestic banks create each economy’s supply of domestic and foreign currency. The model combines the UIP-based and monetary theories of exchange rate determination, but the latter with a focus on private rather than public money creation. The model features an endogenous monetary spread or excess return in the UIP condition. This spread experiences sizeable changes when shocks affect the relative supplies (of bank loans) or demands (for bank deposits) of the two currencies. Under such shocks, monetary effects dominate traditional UIP effects in the determination of exchange rates and allocations, and this becomes stronger as domestic and foreign currencies become more imperfect substitutes. With these shocks, the model successfully addresses the UIP puzzle, and it is also consistent with the Meese-Rogoff and PPP puzzles.
A Procedure for Combining Zero and Sign Restrictions in a VAR-Identification Scheme
(with Alex Haberis)
CFM Discussion Paper 1410 (permanent working paper)
We study exchange rate determination in a 2-country model where domestic banks create each economy’s supply of domestic and foreign currency. The model combines the UIP-based and monetary theories of exchange rate determination, but the latter with a focus on private rather than public money creation. The model features an endogenous monetary spread or excess return in the UIP condition. This spread experiences sizeable changes when shocks affect the relative supplies (of bank loans) or demands (for bank deposits) of the two currencies. Under such shocks, monetary effects dominate traditional UIP effects in the determination of exchange rates and allocations, and this becomes stronger as domestic and foreign currencies become more imperfect substitutes. With these shocks, the model successfully addresses the UIP puzzle, and it is also consistent with the Meese-Rogoff and PPP puzzles.
Work in progress
Fiscal nowcasting
(with Jacopo Cimadomo, Domenico Giannone, Michele Lenza and Francesca Monti)