When the economy rebounds, banks will need the agility to respond quickly and confidently. The models he’s working on today will help them chart a course to profitability, post-recession.

Nomis is first and foremost a data science and analytics company, and so the people who are behind that data are what make this company tick. This week, we set down with Brian Poi, who recently joined Nomis as a data scientist, to learn how he spends his time.

 

Christa:

Your last name struck me as quite unique. What’s the history behind that?

Brian:

My grandparents came to America in the early 1900s from Hungary. Interestingly, from we can tell, the name Poi is apparently not Hungarian. We're not 100% sure where the name came from, but a cousin did some kind of DNA test and found out that we are descendants of Genghis Kahn, leader of the Mongolian Empire. However, it turns out that Poi is in fact a common name in Mongolia, so maybe that’s true. My colleagues will be pleased to know I have inherited few of his personal traits.

Christa:

Fascinating! So how does an economist descended from an empire builder stay busy?

Brian:

We’re developing a new product focused in bank deposits that aims to describe and predict how consumers react to changes in deposit interest rates, among other things. I’m working on developing a core model that will explain consumer behavior with respect to deposits, so most of my day is spent either working with data or working with models to predict that data.

At its core, economics focuses on explaining the behavior of people. In microeconomics we ask questions about individuals: What can I buy? What do I spend my money on? How do I earn a living? In macroeconomics we’re trying to understand and model how all of your and my decisions, taken together, affect the nation’s economy as a whole.

Christa:

Has economics been at the center of your career? Or is your training more in software development?

Brian:

I got my PhD in economics from the University of Michigan. My first job was with a statistical software company called Stata. The software we were developing really requires a PhD in the field to shape its features and functionality. I also spent a good chunk of my time helping our customers apply the software to solving their problems, which exposed to me to the practical application of our technology, which is essential to improving its capabilities.

From there, I went to Moody's Analytics to work on economic forecasting. I spent a lot of time working with banks on consumer credit modeling, various types of forecasting, and stress testing. Eventually, I set up my own consulting firm, which is how I met Nomis.

Christa:

So that’s a full circle back to a software company. What drew you to Nomis?

Brian:

I like building models that help chart a roadmap for banks, and of course that’s a core area of focus for Nomis. I think we can really help them during this challenging economy. There’s no doubt that it will be tough for the next four to six quarters. But eventually the economy will recover, the industrial sector will pick up, and the Fed will start raising the federal funds rate. Banks need to know how to react to that.

One of the fundamental issues of banking is you don't want to pay more than you have to for deposits, but at the same time you definitely don't want to get caught in a position where you have insufficient deposits relative to where you would like to be. The same concept applies to loans; there's always that sweet spot. If you price too low, suddenly you have all these low-interest loans on your books.

Right now, we’re modeling scenarios where the fed interest rate is set to zero for a sustained period and analyzing how banks should prepare for the next set of interest rate hikes in 2021-22.

Christa:

Is it too soon to have a recommendation for banks, given what you’re seeing

Brian:

I would say, especially now, you have to keep your eyes open at all times. We are in uncharted territory — we’ve not seen an economic environment caused by a virus since 1918. The economist in me says that there will always be these Black Swan events that cause a dramatic turn of events. This downturn completely blind-sided us. But, similarly, the next sudden event may take us equally by surprise — the economy may turn around much faster than we expect. That means you’ve got to be vigilant, be continually thinking way ahead, and have contingency plans in place and action plans built on accurate models that can be swiftly implemented. That’s where we come in!

Christa:

Thanks Brian. We’ll check back in with you in a few months.

Take the customer centric survey