In this work, the authors create machine learning (ML) models to forecast home equity credit risk for individuals using a real-world dataset and demonstrate methods to explain the output of these ML models to make them more accessible to the end user. They analyze the explainability for various stakeholders: loan companies, regulators, loan applicants, and data scientists, incorporating their different requirements with respect to explanations. For loan companies, they generate explanations for every model prediction of creditworthiness. For regulators, they perform a stress test for extreme scenarios. For loan applicants, they generate diverse counterfactuals to guide them with steps toward a favorable classification from the model. Finally, for data scientists, they generate simple rules that accurately explain 70%–72% of the dataset. Their study provides a synthesized ML explanation framework for all stakeholders and is intended to accelerate the adoption of ML techniques in domains that would benefit from explanations of their predictions.

We provide a review of macro-finance models featuring nonlinear dynamics. We survey the models developed recently in the literature, including models with amplification effects of financial constraints, models with households' leverage constraints, and models with financial networks. We also construct an illustrative model for those readers who are unfamiliar with the literature. Within this framework, we highlight several important limitations of local solution methods compared with global solution methods, including the fact that local-linearization approximations omit important nonlinear dynamics, yielding biased impulse-response analysis.

We represent risk factors as sums of orthogonal components capturing fluctuations with cycles of different length. The representation leads to novel spectral factor models in which systematic risk is allowed—without being forced—to vary across frequencies. Frequency- specific systematic risk is captured by a notion of spectral beta. We show that traditional factor models restrict the spectral betas to be constant across frequencies. The restriction can hide horizon-specific pricing effects that spectral factor models are designed to re- veal. We illustrate how the methods may lead to economically meaningful dimensionality reduction in the factor space.

We provide a critical review of macroeconomic models used for monetary policy at central banks from a finance perspective. We review the history of monetary policy modeling, survey the core monetary models used by major central banks, and construct an illustrative model for those readers who are unfamiliar with the literature. Within this framework, we highlight several important limitations of current models and methods, including the fact that local-linearization approximations omit important nonlinear dynamics, yielding biased impulse-response analysis and parameter estimates. We also propose new features for the next generation of macrofinancial policy models, including: a substantial role for a financial sector, the government balance sheet and unconventional monetary policies; heterogeneity, reallocation, and redistribution effects; the macroeconomic impact of large nonlinear risk-premium dynamics; time-varying uncertainty; financial sector and systemic risks; imperfect product market and markups; and further advances in solution, estimation, and evaluation methods for dynamic quantitative structural models.

We develop a new cryptographic platform called SCRAM (Secure Cyber Risk Aggregation and Measurement) that allows multiple entities to compute aggregate cyber-risk measures without requiring any entity to disclose its own sensitive data on cyberattacks, penetrations, and losses. Using the SCRAM platform, we present results from two computations in a pilot study with six large private-sector companies: (1) benchmarks of the adoption rates of 171 critical security measures and (2) links between monetary losses from 49 security incidents and the specific sub-control failures implicated in each incident. These results provide insight into problematic cyber-risk-control areas that need additional scrutiny and/or investment, but in a completely anonymized and privacy-preserving way.

Advances in financial technology have made tax-loss harvesting more feasible for retail investors than such strategies were in the past. We evaluated the magnitude of this “tax alpha” with the use of historical data from the CRSP monthly database for the 500 securities with the largest market capitalizations from 1926 to 2018. Given long-term and short-term capital gains tax rates of 15% and 35%, respectively, we found that a tax-loss-harvesting strategy yielded a before-transaction-cost tax alpha of 1.08% per year for our sample period. When the strategy was constrained by the “wash sale rule,” the tax alpha decreased from 1.08% per year to 0.82% per year.

The tenth anniversary of the disastrous weekend that nearly brought down the global financial system is fast approaching. But in many of the jurisdictions that were central to the crisis, financial regulations introduced in the aftermath, aimed at preventing a repeat, are now being rolled back. The pendulum of regulation is now swinging back towards fewer and looser restrictions – and if the past is any guide, a ramp-up in systemic risk exposures will be the result.

August is typically when Wall Street goes to the beach, the mountains, or just home to recharge for a week or two. Many Europeans take the entire month off. But financial markets have a cruel knack of ruining holidays. As we lie in our hammocks this August, we might do well to recall a remarkable event that occurred, seemingly without warning, 11 years ago this month in the run-up to the financial crisis.

The information revolution has transformed everyday life for billions of people throughout the world. For example, according to mobile phone research group GSMA Intelligence, there are currently over 5 billion unique mobile phone subscribers, out of an estimated global population of 7.6 billion. This is the equivalent of a mobile phone for every person on the planet between the ages of 15 and 65.

Stop-loss strategies are commonly used by investors to reduce their holdings in risky assets if prices or total wealth breach certain pre- specified thresholds. We derive closed-form expressions for the impact of stop-loss strategies on asset returns that are serially correlated, regime switching, and subject to transaction costs. When applied to a large sample of individual U.S. stocks, we show that tight stop-loss strategies tend to under-perform the buy-and-hold policy in a mean-variance frame work due to excessive trading costs. Outperformance is possible for stocks with sufficiently high serial correlation in returns. Certain strategies succeed at reducing downside risk, but not substantially.

Breakthroughs in computing hardware, software, telecommunications, and data analytics have transformed the financial industry, enabling a host of new products and services such as automated trading algorithms, crypto-currencies, mobile banking, crowdfunding, and robo-advisors. However, the unintended consequences of technology-leveraged finance include firesales, flash crashes, botched initial public offerings, cybersecurity breaches, catastrophic algorithmic trading errors, and a technological arms race that has created new winners, losers, and systemic risk in the financial ecosystem. These challenges are an unavoidable aspect of the growing importance of finance in an increasingly digital society. Rather than fighting this trend or forswearing technology, the ultimate solution is to develop more robust technology capable of adapting to the foibles in human behavior so users can employ these tools safely, effectively, and effortlessly. Examples of such technology are provided.

The authors introduce a new approach to identifying and monitoring systemic risk that combines network analysis and tail risk contribution (TRC). Network analysis provides great flexibility in representing and exploring linkages between institutions, but it can be overly general in describing the risk exposures of one entity to another. TRC provides a more focused view of key systemic risks and richer financial intuition, but it may miss important linkages between financial institutions. Integrating these two methods can provide information on key relationships between institutions that may become relevant during periods of systemic stress. The authors demonstrate this approach using the exposures of money market funds to major financial institutions during July 2011. The results for their example suggest that TRC networks can highlight both institutions and funds that may become distressed during a financial crisis.

Using account level credit-card data from six major commercial banks from January 2009 to December 2013, we apply machine-learning techniques to combined consumer-tradeline, credit-bureau, and macroeconomic variables to predict delinquency. In addition to providing accurate measures of loss probabilities and credit risk, our models can also be used to analyze and compare risk management practices and the drivers of delinquency across the banks. We find substantial heterogeneity in risk factors, sensitivities, and predictability of delinquency across banks, implying that no single model applies to all six institutions. We measure the efficacy of a bank’s risk-management process by the percentage of delinquent accounts that a bank manages effectively, and find that efficacy also varies widely across institutions. These results suggest the need for a more customized approached to the supervision and regulation of financial institutions, in which capital ratios, loss reserves, and other parameters are specified individually for each institution according to its credit-risk model exposures and forecasts.

Technological advances in telecommunications, securities exchanges, and algorithmic trading have facilitated a host of new investment products that resemble theme-based passive indexes but which depart from traditional market-cap-weighted portfolios. I propose broadening the definition of an index using a functional perspective—any portfolio strategy that satisfies three properties should be considered an index: (1) it is completely transparent; (2) it is investable; and (3) it is systematic, i.e., it is entirely rules-based and contains no judgment or unique investment skill. Portfolios satisfying these properties that are not market-cap-weighted are given a new name: “dynamic indexes.” This functional definition widens the universe of possibilities and, most importantly, decouples risk management from alpha generation. Passive strategies can and should be actively risk managed, and I provide a simple example of how this can be achieved. Dynamic indexes also create new challenges of which the most significant is backtest bias, and I conclude with a proposal for managing this risk.

In this opinion piece, MIT Sloan Professor Andrew Lo and Princeton Professor Simon Levin write, "We propose that the financial system has crossed a threshold of complexity where the system is evolving faster than regulators and regulations can keep pace. For example, the system is now truly globally connected, but coordination across sovereign jurisdictions is difficult to achieve. This new situation calls for a new perspective, one based on a different paradigm than the ones on which financial regulation is currently based, such as efficient markets, rational expectations, and models patterned after the physical sciences."