« Back to Results

Navigating Uncertainty and Price Dynamics in Modern Economies

Paper Session

Sunday, Jan. 4, 2026 2:30 PM - 4:30 PM (EST)

Philadelphia Marriott Downtown, Room 411
Hosted By: Society for Nonlinear Dynamics and Econometrics & Society for Nonlinear Dynamics and Econometrics
  • Chair: Yoosoon Chang, Indiana University-Bloomington

Dual Interpretation of Machine Learning Forecasts

Philippe Goulet Coulombe
,
Université du Québec à Montréal
Maximilian Göbel
,
Brain
Karin Klieber
,
Oesterreichische Nationalbank

Abstract

Machine learning predictions are typically interpreted as the sum of contributions
of predictors. Yet, each out-of-sample prediction can also be expressed as a linear
combination of in-sample values of the predicted variable, with weights corre-
sponding to pairwise proximity scores between current and past economic events.
While this dual route leads nowhere in some contexts (e.g., large cross-sectional
datasets), it provides sparser interpretations in settings with many regressors and
little training data—like macroeconomic forecasting. In this case, the sequence
of contributions can be visualized as a time series, allowing analysts to explain
predictions as quantifiable combinations of historical analogies. Moreover, the
weights can be viewed as those of a data portfolio, inspiring new diagnostic mea-
sures such as forecast concentration, short position, and turnover. We show how
weights can be retrieved seamlessly for (kernel) ridge regression, random forest,
boosted trees, and neural networks. Then, we apply these tools to analyze post-
pandemic forecasts of inflation, GDP growth, and recession probabilities. In all
cases, the approach opens the black box from a new angle and demonstrates how
machine learning models leverage history partly repeating itself.

Multinationals and Uncertainty: The Role of Internal Capital Markets

Jianlin Wang
,
Leavey School of Business, Santa Clara University

Abstract

Despite their importance, we know little about how the internal capital markets
(ICMs) of multinational enterprises (MNEs) connect external lenders. My paper ad-
dresses this gap by studying the interaction between parent- and subsidiary-level external lenders. I first derive the optimal debt structure of MNEs, explaining the role of
subsidiary-level external debt. When used to incentivize delegated monitoring, I show
that MNEs can substitute it with cheaper parent-level external debt to offset delever-
aging pressure from local shocks that enhance the monitoring incentives. Using Brexit
uncertainty as a natural experiment, I provide evidence of this channel at both parent
and subsidiary levels.

Divergent Perceptions, Divergent Pay: Inflation and the Gender Wage Gap

Lovisa Reiche
,
BI Norwegian Business School
Nicolo' Maffei-Faccioli
,
Norges Bank

Abstract

How does the gender wage gap respond to inflation? We show that it widens after both supply- and demand-driven inflationary shocks. This widening reflects gender differences in labor market perceptions: women interpret inflation as a signal of deteriorating conditions, while men perceive mild improvement. These divergent beliefs reduce women’s willingness to negotiate, slowing their nominal wage growth relative to men’s. To formalize this mechanism, we develop a New Keynesian search-and-matching model where workers do not observe the true nature of the shock and women form ambiguity averse beliefs while men are ambiguity loving. The model replicates the observed cyclicality of the gender wage gap, establishing a novel link between inflation and gender inequality.

Debt Indexation, Determinacy, and Inflation

Tobias Kawalec
,
University of Oxford, Department of Economics and Nuffield College

Abstract

Contrary to popular belief, inflation-indexed government debt can boost inflation in response to deficit shocks, conditional on a lack of sufficient future fiscal backing. I formalize this insight through a state-of-the-art calibrated HANK model with multiple asset types, showing that the annual inflationary effect of a 1% deficit-to-GDP shock increases by 0.35 percentage points when 30% of the government debt stock is indexed to inflation, as is the case in the United Kingdom, relative to a baseline case calibrated to the United States. Inflation-indexed debt makes the price level partially backward-looking through the government debt valuation equation, thereby causing additional inflationary pressure. Empirical evidence from a large, narratively identified fiscal deficit shock supports this finding, which has additional implications for the distinction between ’fiscally-led’ mechanisms and ’HANK-type’ mechanisms surpassing Ricardian equivalence.

Discussant(s)
Ulrich K. Müller
,
Princeton University
Dejanir H. Silva
,
Purdue University
Hie Joo Ahn
,
Federal Reserve Board
Margaret M. Jacobson
,
Federal Reserve Board
JEL Classifications
  • E3 - Prices, Business Fluctuations, and Cycles
  • C5 - Econometric Modeling