As I reflect on my research contributions over the past 40+ years, they fall into five phases. In what follows, I briefly describe each phase as well as a few of the significant papers in each area. What unifies all this work is the desire to address practical problems using methodologies from Management Science and Operations Research. This led, in some cases, to the development of new methodologies to address these problems, which have stood the test of time in their contributions to the field.

Because the fifth phase is where my active work now sits, I begin there. The four earlier phases follow in chronological order.

Current Work — Research Phase V: Payments and Financial Market Infrastructure

2026–present

The financial system has become a real-time system running on shared infrastructure. Money now moves within seconds, around the clock, with finality; and the institutions that move it increasingly depend on a small common layer of cloud platforms, core processing vendors, payment and market utilities, data and identity services, and a growing layer of artificial-intelligence model providers. Both shifts are recent, both are accelerating, and both change the risk profile of the system in ways the existing regulatory architecture was not designed to address. My current research asks what this architecture does to systemic risk and what supervisors and operators should do about it. The work proceeds along three related lines.

Speed, fraud, and liquidity in real-time payment systems

Instant settlement is not free. By eliminating settlement delay, real-time systems disable two mechanisms that had quietly stabilized the payment system for decades: the window in which fraud can be screened, and the interbank netting that economizes on liquidity. In “Speed, Fraud, and Bank Liquidity” I model this tradeoff directly and show that the welfare-optimal design requires a positive settlement delay — measured in seconds, and roughly double what fraud considerations alone would justify, once the liquidity effects on banks are taken into account. The analysis incorporates suspension mechanisms analogous to market circuit breakers and draws on the operating experience of Brazil’s Pix system.

A companion paper, “The Tightening Trap,” takes up the fraud problem as an operational one rather than a purely statistical one. Fraud detection is not a classifier operating in isolation; it is a queueing system with finite review capacity, embedded in a network of institutions that route transactions among themselves. Using that framing, I show that tightening detection at a single institution can paradoxically increase aggregate losses across the system, and that detection-only regulatory mandates can backfire when they push review operations toward their capacity limits without corresponding investment in operational resilience.

Concentration in critical infrastructure and systemic fragility

As institutions outsource more of their information systems to a common technology layer, the concentration of that layer itself becomes a source of system-level risk. In “Critical Infrastructure Concentration and Systemic Financial Fragility” I develop an analytical model in which a continuum of financial institutions chooses provider portfolios from a market with endogenous concentration, and in which providers fail through both idiosyncratic and common-cause channels. The central result is that equilibrium diversification falls strictly short of the social optimum whenever a joint-failure externality is present, with provider concentration strictly above the planner’s level when entry responds. Following Weitzman (1974), the paper compares a price on concentration against a quantity rule requiring minimum portfolio diversity, and finds that the quantity rule typically dominates under empirically plausible conditions, because the marginal social damage from concentration is steep near the concentrated region while the marginal private cost of diversification is comparatively flat. A robust-optimization characterization of optimal portfolios addresses the case in which the common-cause failure structure is itself uncertain. A separate technical companion collects the general-model counterparts and proofs.

The design of operational resilience regulation

The third line translates the analysis into terms a supervisor can use. “Who Manages the Variance?” organizes the policy problem around a single identity: expected systemic damage from operational failure is governed by a mean and a variance. The market already manages the mean, because every institution cares about its own uptime. Nobody manages the variance, which is determined by who fails together — and concentration, cross-layer alignment, and the timing of failures are all variance, not mean. Five recommendations follow: cap system-wide concentration rather than mandating firm-level diversity alone; prefer quantity instruments to price instruments; calibrate the instrument to each provider market; mandate dependency disclosure and build the system-level map; and pair any concentration instrument with reliability standards. The paper confronts the strongest practical objection — that in some markets, above all core processing, no institution can realistically run two vendors — and shows that the framework’s answer is not to force dual sourcing but to shift the instrument.

This work is timely as well as analytical. In July 2026 the United Kingdom placed four cloud providers under direct supervisory oversight as its first designated critical third parties — the first concrete instance of the oversight architecture these models evaluate — and comparable regimes are being implemented by the Financial Stability Board, the European Union, and national regulators.

Why this work, and why now

This phase returns me to the kind of modeling with which I began: equilibrium analysis, queueing, and optimization under uncertainty, now applied to the plumbing of the financial system rather than to freight networks. The questions, however, came directly from the decade I spent at the Federal Reserve Bank of Philadelphia, where the operational architecture of payments and the concentration of the vendors supporting it were live supervisory concerns long before they were literatures. It is, in that sense, the most direct expression of what has motivated the whole of this work: a practical problem, taken seriously enough to require a model.

Representative papers

• P.T. Harker, “The tightening trap: detection, capacity and strategic routing in multi-institution fraud prevention networks,” Working paper, Department of Operations, Information and Decisions, The Wharton School, University of Pennsylvania, April 2026. https://dx.doi.org/10.2139/ssrn.6563339
• P.T. Harker, “Speed, fraud, and bank liquidity: the operational architecture of real-time payment systems,” Working paper, Department of Operations, Information and Decisions, The Wharton School, University of Pennsylvania, August 2026. https://dx.doi.org/10.2139/ssrn.7270940
• P.T. Harker, “Critical infrastructure concentration and systemic financial fragility,” Working paper, Department of Operations, Information and Decisions, The Wharton School, University of Pennsylvania, August 2026. https://dx.doi.org/10.2139/ssrn.7282881
• P.T. Harker, “Critical infrastructure concentration and systemic financial fragility: general model and proofs (technical companion note),” Working paper, Department of Operations, Information and Decisions, The Wharton School, University of Pennsylvania, August 2026. https://dx.doi.org/10.2139/ssrn.7282120
• P.T. Harker, “Who manages the variance? Operational concentration and the regulation of financial infrastructure,” Working paper, Department of Operations, Information and Decisions, The Wharton School, University of Pennsylvania, August 2026. https://dx.doi.org/10.2139/ssrn.7285539

Related presentation

• “The Architecture of Real-Time Payments,” Federal Reserve System Payments Researchers Annual Meeting, Atlanta, GA, August 2026.

The Earlier Phases

Research Phase Ia: Transportation Science

The Staggers and Motor Carrier Acts of 1980 were major pieces of legislation that deregulated the transportation industry in the United States. Starting with my dissertation work, which was funded by the U.S. Department of Energy, a series of articles were written concerning how these pieces of landmark legislation would change our freight transportation system. These articles describe both the theory of and algorithms for very complex network models of competition. The models were then used by the Department of Energy and other federal agencies to predict freight flows both here in the United States and abroad. Three representative examples of this literature are:

• T. Harker and T.L. Friesz, “Prediction of intercity freight flows I: theory,” Transportation Research 20B (1986), 139–153.
• T. Harker and T.L. Friesz, “Prediction of intercity freight flows II: mathematical formulations,” Transportation Research 20B (1986), 155–174.
• T. Harker, “Alternative models of spatial competition,” Operations Research 34 (1986), 410–425.

Research Phase Ib: Variational Inequality and Complementarity Theory and Algorithms

The models developed in Phase Ia required new approaches to solving very complex and large-scale computable equilibrium problems. As a result, I undertook with my colleague Jong-Shi Pang and several of my doctoral students a multi-year effort to develop both the theory and the algorithms for this class of problems. These were formulated as variational inequality or complementarity problems. This work led to a significant number of advances in the field that are best summarized in an article I co-authored with Jong-Shi Pang, which not only reviewed the advances in the field but also closed a few of the holes in the theory that we discovered during the process of writing it:

• T. Harker and J.S. Pang, “Finite-dimensional variational inequality and nonlinear complementarity problems: a survey of theory, algorithms and applications,” Mathematical Programming 48 (1990), 161–220.

To this day this article is seen as the main reference for the field and continues to be cited 35 years after its publication.

Research Phase II: Optimization Models for Freight Transportation Systems

Soon after joining the Wharton faculty, I was awarded a Presidential Young Investigator grant from the National Science Foundation. This award not only recognized excellence in one’s early career trajectory but also encouraged university–industry collaboration. As a result, I received funding from the Burlington Northern Railroad to help develop the optimization models that became part of the Advanced Railroad Electronics System, or ARES. This work again led to several doctoral students working on the project as well as to patents on the underlying algorithms. Two examples of this work include:

• Kraay, P.T. Harker and B. Chen, “Optimal pacing of trains in freight railroads: model formulation and solution,” Operations Research 39 (1991), 82–99.
• Jovanovic and P.T. Harker, “Tactical scheduling of rail operations: the SCAN~I decision support system,” Transportation Science 25 (1991), 46–64.

Research Phase III: Service Operations and Financial Services

Having become the Director of the Fishman Davidson Center for the Study of the Service Sector at Wharton, I then pivoted my work to broader issues related to the service economy and, specifically, financial services. This work was supported by a major grant from the Sloan Foundation that we received in the Financial Institutions Center, where I led the productivity study in retail banking. Once again, a group of graduate students joined me in conducting this research, which led to a stream of work related to competition in the service sector, productivity in financial services, and broader concepts of customer efficiency and its management. A few representative articles in this space are:

• T. Harker and S.A. Zenios, “What drives the performance of financial institutions?” in P.T. Harker and S.A. Zenios (eds.), Performance of Financial Institutions (Cambridge University Press, London, 2000), 3–31.
• P. Cachon and P.T. Harker, “Competition and outsourcing with scale economies,” Management Science 48 (2002), 1314–1333.
• Xue and P.T. Harker, “Customer efficiency: concept and its impact on e-business management,” Journal of Service Research 4 (2002), 253–267.
• Xue, L.M. Hitt and P.T. Harker, “Customer efficiency, channel usage and firm performance in retail banking,” Manufacturing and Service Operations Management 9 (2007), 535–558.
• X. Frei, P.T. Harker and L.W. Hunter, “Retail banking,” in D.C. Mowery (ed.), U.S. Industry in 2000: Studies in Competitive Performance (National Academy Press, Washington, DC, 1999), 179–214.

Research Phase IV: Dean, University President and Federal Reserve President

The next pivot I made was to begin a career as an academic leader and, for the last decade, as President and CEO of the Federal Reserve Bank of Philadelphia. While my ability to publish research was and is limited in these roles, I continued to invest in state-of-the-art research capability in each institution. Thus, I moved from being the author of the work to developing the researchers and the infrastructure to make significant advances in a wide variety of disciplines. That said, as Fed President we still communicated our ideas, but in the form of speeches, which can be found here:

https://www.philadelphiafed.org/search-results/all-work?searchtype=speeches-harker

It was also in this period that the questions animating the fifth phase of my work took shape. The operational architecture of the payment system and the concentration of the technology providers on which financial institutions depend were supervisory concerns at the Fed well before they became research literatures — and returning to Wharton gave me the opportunity to model them properly.

My teaching draws on the two threads that have defined my career: the operations and technology of financial institutions, and the practice of monetary policy and central banking. In the MBA classroom, I teach OIDD 6750, Financial Services Management, which treats every financial institution—from the largest global bank to the newest fintech—as, at its core, an operating system, and examines how these firms design, deliver, and protect the processes through which financial services reach customers. The course is built around three questions: how do you design a financial service that works, how do you keep it from breaking, and how do you make it materially better? It closes with what I call stability-induced fragility—the idea that the very tools we deploy to make the financial system more stable can, at scale, generate new vectors of failure. Throughout, I pair quantitative methods (queuing theory, capacity modeling, process analysis) with technology depth (AI systems, payment networks, cybersecurity, cloud infrastructure) and the practitioner and regulatory perspective I gained over a decade leading the Federal Reserve Bank of Philadelphia.

That same vantage point shapes my executive education teaching. In programs such as How Monetary Policy Is Really Done: An Insider’s View, I take participants inside the FOMC—the intermeeting period and the intelligence the Fed sees that markets don’t, the dynamics of the go-around, how a single word in the policy statement can represent hours of negotiation, and the judgments about unobservable variables like the neutral rate (r*) that underlie every decision—connecting the mechanics of central banking to the cost of capital, valuations, and strategic choices that executives and investors actually face.

All Courses

• OIDD6750 - Financial Services Management

  The financial services industry is a foundational pillar of both the U.S. and global economies, providing critical infrastructure for capital markets, credit intermediation, risk management, and financial planning. In the United States, the sector accounts for approximately 7.5% of GDP (excluding real estate), contributing over $2.2 trillion annually, and employs more than 8.5 million people. This course explores how process and technology management can be applied to improve efficiency, productivity and quality of this sector. We will take the traditional process view of operations to study how to design, deliver and innovate financial service processes with advances in technology. Drawing on the lessons from research and best practices, and from the perspective of financial institutions and regulators, the following topics will be covered: Designing Financial Services, Operational Risk Management in Financial Services, Innovating Financial Services. The course content blends rigorous quantitative and qualitative analysis, theories and practices, and practitioner and regulatory perspectives. The goal of this course is to cultivate the skills mandated by today’s increasingly complex and challenging environment facing financial services organizations.

• The Lawrence C. Nussdorf Urban Leadership Prize, Penn Institute for Urban Research, University of Pennsylvania, April 2025
• The James B. O’Neill Award for Excellence in Economic Education and Entrepreneurship, University of Delaware, February 2025
• Honorary Doctor of Science, Thomas Jefferson University, May 2017
• Honorary Doctor of Science, University of Delaware, May 2016
• D. Robert Yarnell Award (Engineering Alumnus of the Year), School of Engineering and Applied Sciences, University of Pennsylvania, May 2016
• Order of the First State awarded by Governor Jack Markell, May 2015
• Charter Fellow, National Academy of Inventors, December 2012
• Fellow, Institute for Operations Research and the Management Sciences (INFORMS), October 2012
• Monsignor Thomas Reese Award, Catholic Charities of the Diocese of Wilmington, Delaware, recognizing exemplary individuals who have demonstrated a deep commitment to promoting and restoring the well-being of people – Catholic Charities’ mission.
• Outstanding Advocacy of Small Business, U.S. Small Business Administration (Delaware), June 2011
• Honorary Degree, Xiamen University, China, October 2009
• Joseph Wharton Award, Wharton Club of Washington, DC, November 2008
• Ivy Football Association honoree, January 2007
• Omega Rho Distinguished Lecturer, INFORMS, November 2006
• 2002 Alan Goldman Lecturer, Department of Mathematical Sciences, The Johns Hopkins University
• 2002 Wei Lun Distinguished Visiting Professor, Chinese University of Hong Kong
• 1998 David W. Hauck Award for Outstanding Teaching in the Undergraduate Division, The Wharton School
• Laurent Picard Distinguished Lecturer, McGill University, Montreal Canada, 1998
• CORE Lecturer, Center for Operations Research and Econometrics, Universite Catholique de Louvain, Belgium, 1993
• 1992 Miller-Sherrerd MBA Core Teaching Award, The Wharton School
• White House Fellow, 1991–92 (named by President George H. W. Bush)
• National Science Foundation Presidential Young Investigator Award, 1986–91on

• Looking back at Jerome Powell’s 8-year term as Federal Reserve Chair, National Public Radio - 05/10/2026
• The Fed cannot fix what it did not break, The Hill - 03/15/2026

Knowledge at Wharton

• The Fed Explained: What It Does and Why It Matters, Knowledge at Wharton - 7/7/2026
• Federal Reserve Leadership Transition Amid Economic and Political Pressures, Knowledge at Wharton - 5/27/2026
• Why Interest Rates Can’t Fix Deeper Economic Problems, Knowledge at Wharton - 1/28/2026