Blog

Here you will find information about our past and ongoing projects, as well as opinions on current topics in Banking Analytics.

July 22, 2024by Cristian0

It’s rare that I get to speak more about the tech side of my Fintech expertise (beyond AI, of course), but with the CrowdStrike bug making a mess around the world, I had the chance to do so on Saturday at CBC and Friday at CTV.

We discussed the impact of this on the financial system, and potential future measures to take given this critical infrastructure’s vulnerability. My take here is that we don’t put the same level of scrutiny to software companies as we do physical infrastructure, even though the consequences of a severe outage of the former can be just as severe. We trust that the companies themselves will have teams that do sufficient quality control. Why? For most critical systems, we require independent verifications. Why not for these companies that supervise, with full privileged access, a big chunk of the corporate systems?

Watch the Weekend Business Panel at the link below!



July 17, 2024by Cristian0

I was asked my opinion on the latest inflation numbers today by Global News. I got to try Western University‘s new studio! Green screen and all.

In summary: Almost in line with what was expected, 2.7% vs 2.8%. There is an 86% chance of a cut next week, but I doubt it will be more than 0.25%, too early to stimuli the economy too much.

The lighting was a bit too strong, so I had to take off my glasses xD. Give it a watch at the link below!

Bank of Canada rate-cut odds rise after June inflation release


Scale-e1720480798699.png?fit=512%2C512&ssl=1

July 8, 2024by Cristian0

This is a bit late, but last month I was very popular after the interest rate decrease by the BoC. Western asked me to write an explainer, which they told me was the most visited article at Western News! I am copying it here for posterity. I also beat a personal record: I had five interviews in 24 hours. It was really a topic that garnished a lot of attention! I appeared in the London Free Press, The X, CHCH News, CFPL News, and Global Radio.

It was great to see that I nailed my prediction too. For July, the rate should remain stable. Too early to tell the consequences of the economy of the very first one. I fully expect a decrease in the September announcement, though. The explainer follows:

Western News: Can Canadians expect an interest rate cut?

Cristián Bravo: Given the latest downward trend on inflation and economic growth (the production of goods and services in an economy), the idea of a rate cut is much more likely. We are seeing a generalized cooling down in the economy that has been persistent over a nine-month period, signalling that the efforts by the Bank of Canada have been successful. The fact that growth is now lower than expected, makes it more likely that the Bank of Canada will decide to ease on their position and start lowering the rate and seeing how the market reacts.

It will need to balance the potential risk of stimulating the economy too early, thus leading to a return of inflation, versus the chance that the slowing growth trend continues, and we enter a recession, as we seemed to have been on the edge of during the last two quarters of 2023.

Why would the Bank of Canada not cut rates?

CB: What may give the Bank of Canada pause are the numbers in the U.S.

The economy there is still in excess demand and inflation has not eased. There has been significant volatility in the core consumer price index and consumption numbers, meaning that not even the Federal Reserve knows the right path to take. This affects us, as the Canadian rate and the U.S. rate cannot diverge too much, or the Canadian dollar will lose value significantly against the U.S. dollar, undoing some of the efforts of the Bank of Canada.

The Bank of Canada does have leeway to lower the rate but needs to be cautious because if the U.S. decides to keep rates high for a while, then we won’t be able to lower them at a higher speed. So, a moderate decrease of 25 basis points (or 0.25 per cent) is likely, although I wouldn’t be surprised if they decide to keep it at its current value and wait until the July meeting to see how the American economy evolves in relation to our own.

What role does inflation play in monetary policy?

CB: Inflation, growth and employment are the trinity of monetary policy. The Bank of Canada controls, through their policy rate, the price of lending money, and this directly controls how much money is available to go around. Too much money related to our capacity to produce goods and services leads to inflation. Too little, leads to a credit crunch and thus decreased growth.

Jobs are directly tied to this. An overheated economy, or an economy without as many workers as needed (as we had a few months ago), leads to inflationary processes, while an economy in depression leads to job losses as businesses need to adapt to the lower demand for their products and services. So, the Bank of Canada’s role is to set the incentives to either stimulate the economy, or to disincentive spending, and this is done by changing the cost of borrowing funds through monetary policy.

What are the benefits of raising or lowering lending rates?

CB: Lower rates mean more incentive to lend money, and thus to invest, hire, spend and produce more. If this is tied to a real need for those goods and services, then growth happens, salaries increase and employment grows. If there is more money than needed in the economy, and we are observing inflation as we were last year, then a higher rate has the opposite effect, disincentivizing spending and demand. The tricky part is reaching a rate that leads to sustainable levels of employment and spending so that we achieve growth and higher salaries while producing goods and services that are aligned with local and global demand.



May 27, 2024by Cristian0

A new panel is live, and now we are featured on the CBC website! This time we spoke about three topics:

  1. The DoJ suing to split back Ticketmaster / Live Nation: The most significant change versus previous attempts is that now they managed to get everyone angry. Artists, venues and consumers are all affected. Oddly, due to weak antitrust laws in the US, consumers are not considered direct customers of Ticketmaster, so now that artists and venues are shown to be affected, the DoJ could intervene. If successful, this should increase competition in the artist/venue space, not in the consumer one, so don’t expect ticket prices to come down anytime soon.
  2. The new low fare from WestJet: WestJet is moving towards an Ultra Low-Cost Carrier (ULCC) model, charging now for the carry-on baggage. This is common in European ULCC, such as EasyJet and others. Changes like this are great for the company as they save on fuel, save on airport fees and taxes, and provide marginally better customer service to those who require overhead bins as there is less demand for them. It also comes with a modest 2% decrease in fares in competitive routes. However, the cost is more fare dispersion. Now you won’t really know what the final price of your ticket will be, and you will need to plan a bit ahead to what you want to bring with you. Also, airports may need to rethink their fee structure, considering they are normally tied to the base price of the ticket.
  3. Inflation: The latest inflation numbers were positive for Canada, with a 2.7% annualized rate. However, and what I spoke about in the panel, I expect the BoC will be a bit wary of being too aggressive lowering the interest rate, given the US is still suffering high inflation and the rates between the two countries cannot diverge too much.

Give it a watch and comments always welcome!


20240330-WBP-scaled.jpg?fit=1200%2C676&ssl=1

April 1, 2024by Cristian0

I was at the CBC News’ Weekend Business Panel this week, speaking about some interesting news that happened. Sadly, the CBC changed their policies, and now we don’t get a video of our participation, so I will be publishing these short summaries after every time I appear. This week we spoke of:

  • The MLS judgment in the US that may change the incentive structure of realtors: Looking at this from a pure incentive structure, the realtor business is poorly constructed. The buyer pays commission to their realtor based on a percentage of the purchase price, which means there are no economic incentives for their realtor to get them their best price (although they do have a fiduciary duty).
    • The lawsuit in the US will change the structure of the process. It will now require a contract between the realtor and the buyer directly, with agreed fees, before showing houses. Buyer representation agreements are already common in Canada, this lawsuit splits the buyer and the seller’s commission, thus providing incentives to realtors to lower their fees when representing the buyer.
      • There is a new rule prohibiting offers of broker compensation on the MLS, and from creating rules that would permit a seller’s agent to determine compensation for a buyer’s agent.
      • However, this also means homebuyers will have to consider an extra closing price, instead of the now baked into the mortgage fee. Fees should come down but will also need to be paid up front.
      • In the US, the realty companies are saying they will not change their practices, as nothing in the judgment forces them to. There is a difference in interpretation on what the judgment actually means. This will most likely lead to new lawsuits if the actual implementations differ for what the other side interpreted.
  • Home Depot’s acquisition of building material supplier SRS Distribution. Home Depot’s thinking is that growth will come from contractors as opposed to retail, that boomed during the pandemic and is now coming down. Their bet is that construction of new homes and government plans to stimulate construction in general will mean higher sales than what they are seeing in their stores.
    • Home Depot said that when taking the deal into account, it now believes its total addressable market is approximately $1 trillion, an increase of approximately $50 billion. Home Depot controls 17% of the market.
    • One pain point in Home Depot has always been logistics, one of SRS’ strengths with their warehouses and truck fleet. This can bring synergies into their main business, even though SRS will continue operating as an independent entity. Through the deal, expected to close by the end of fiscal 2024, Home Depot will add SRS’ network of more than 2,500 professional sales force in 760 plus locations to its footprint of over 2,000 U.S. stores and distribution centres. It would also allow Home Depot to take advantage of SRS’ more than 4,000 truck fleet and job site delivery capabilities.
    • There is still regulatory approval necessary. I am sure Lowe’s will have something to say about this deal. Maybe I’ll get to talk about this later again.
  • Cocoa prices have reached their highest value ever, hitting USD $10,000 per tonne. This is caused by a multifaceted problem. Short term: El Niño and West Africa pests, the swollen-shoot virus and black-pod disease, have been causing havoc with plantations. Just the Swollen-shoot virus affected 20% of all cocoa trees in the Ivory Coast. The war in Ukraine has also caused the sugar prices to go up, thus impacting further the price of chocolate.
    • Long term, though, there is a geopolitical issue. Farmers get about 5% of the price of a bar, or 30% – 50% of the price of a tonne of cocoa. Each producer can make around 1 tonne per year, thus the income of a farmer is around USD $5,000 yearly at best. This has lead to unsustainable practices. 14% of the Ivory Coast and 11.5% of Ghana are cocoa plantations and many are planted in protected areas, 37% of the Ivory Coast and 13% of Ghana’s deforestation comes from cocoa planting.
    • Any solution is super complex. As hard as solving hunger in Africa.  Only a mixture of better governments, better access to sustainable farming training and supplies, less corruption, more development and a strong coordination between governments and international agencies can tackle this. Sadly, to me, this hints we won’t see chocolate prices come down anytime soon, and if the underlying issues are not resolved, we will end up with chocolate scarcity in the long term.

Happy to hear your thoughts about this. I’ll be again next time in May. Always a fun experience!


JetsAwardAAAI.jpg?fit=1200%2C904&ssl=1

March 6, 2024by Cristian0

The Association for the Advancement of Artificial Intelligence (AAAI) conference is recognized as an elite conference in the field of artificial intelligence (AI), gathering a community of academic and industry experts engaged in both theoretical and applied AI research. The 38th AAAI conference was held in the city of Vancouver, Canada, from February 20th to 27th. It offered an excellent opportunity for sharing research work in AI, bridging potential collaborations, and offering the guidance of the direction of AI innovation.

Both Cristian and Jet attended. Jet presented our latest research: “Breaking Barriers: Unveiling Gender Disparities in Corporate Board Career Paths using Deep Learning.” This research was showcased in two workshops: the AI in Finance for Social Impact, organized by J.P. Morgan, and the AI for Financial Services, highlighting the cross-disciplinary impact and relevance of our work. In the former, the lab was also represented in the work “Extreme climate events and credit risk: stress testing approach for loans to SMEs based on network analysis”, presented by the PhD candidate Camilla Carpinelli, part of the group led by Prof. María Óskarsdóttir, at Reykjavík University.

Celebrating Recognition: Best Poster Award

Our collective efforts were recognized with the Best Poster Award at the workshop, a moment of great pride and joy for our team. We extend our deepest gratitude for this honour to the organizers, reaffirming our commitment to contributing meaningful insights and solutions in the realm of AI and social impact in banking.

Networking, Collaboration, and Future Horizons

The AAAI conference provided an excellent opportunity for engagement with the community of AI in Finance. Our team had a great time to meet, network, and exchanging ideas with fellow researchers and practitioners from both academia and industry. Notable interactions included enlightening discussions with peers from prestigious institutions and companies such as CMU, J.P. Morgan, and Capital One, among others.

As we reflect on our experiences at the AAAI conference, we are filled with gratitude for the opportunity to contribute to and partake in the global dialogue on AI. We look forward to attending the next one in 2025!


HouseNetwork-e1707236743536.jpg?fit=200%2C200&ssl=1

February 6, 2024by Cristian0

Our latest preprint, titled “Attention-based dynamic multilayer graph neural networks for loan default prediction”, introduces a novel model that could enhance the accuracy of credit risk assessments.

Credit Scoring and Correlated Default

The inspiration for this work comes from our previous studies, which clearly show that borrowers are not isolated entities, but part of a complex web of connections that can influence their probability of default. This interconnectedness suggests that a borrower’s risk of default may be impacted not just by their financial situation, but also by the network of relationships they are part of.

Our study leverages these insights, proposing a sophisticated model that combines Graph Neural Networks (GNNs) and Recurrent Neural Networks (RNNs) to assess credit risk. This way, the model can use dynamic multilayer networks, each layer reflecting a different source of network connection. The proposed model considers different types of connections between borrowers, such as geographical location and choice of mortgage provider, and the evolution of these connections over time.

How It Works

GNNs are a class of deep learning models designed to operate on graphs — structures that represent relationships between entities. These models are adept at capturing the complex patterns inherent in networks of borrowers. On the other hand, RNNs excel at processing sequential data, making them ideal for analyzing the temporal dynamics of these borrower networks.

The model introduced in our study, by our PhD student Sahab Zandi, in collaboration with Prof. Christophe Mues and Kamesh Korangi from the University of Southampton and Prof. María Óskarsdóttir from Reykjavík University, adds a layer of sophistication with an attention mechanism. This mechanism prioritizes certain time points over others, based on their relevance to the borrower’s default risk. Such an approach allows for a more nuanced analysis, distinguishing the model from traditional methods of credit scoring.

Empirical Evidence of Superior Performance

When tested against a dataset provided by U.S. mortgage financier Freddie Mac, the model not only outperformed traditional credit scoring methods but also offered more in-depth insights into the nature of default risk. We found that the model’s ability to account for the dynamic and multilayered nature of borrower connections enhanced its predictive accuracy. This suggests that the future of credit scoring lies in the ability to understand and model the complex web of relationships that influence financial behaviour.

Looking Ahead

The implications of this study are multifaceted. For lenders, adopting such models could help understand how risk propagates and affects both individual borrowers and their portfolios in a high-stakes market. For borrowers, it could translate into more access to credit by empowering so-called ‘Second Look’ models, which provide thin-file borrowers with a more detailed evaluation. Our results can be part of such evaluation. And for the field of operational research and finance at large, this study paves the way for further exploration into the use of machine learning and network science in multilayered, dynamic, environments.

As we move forward, the exploration of even more sophisticated models — incorporating additional layers to capture a broader array of connections or employing different types of GNNs and RNNs — promises to unlock new insights into credit risk and beyond. The journey towards a more interconnected and intelligent approach to credit scoring is just beginning, and its potential benefits for both lenders and borrowers are immense.

Interested in the topic? Read the working paper on ArXiV!


Untitled150x150px4.png?fit=150%2C150&ssl=1

February 4, 2024by Cristian0

A bit late to the party, but as the first paper on this topic was recently published at EJOR (ArXiV, Journal), I thought it may merit the post.

Last year at the credit scoring conference, I was interviewed by Brendan Le Grange, host of the podcast How to Lend Money to Strangers. We discussed the paper I presented there, part of Sherly’s PhD thesis, on causal models for credit limit settings. The preprint of that paper will appear soon, read the first one meanwhile!

If you want to hear more about my thoughts on credit limit setting and how it relates to causal modelling, listen to the podcast episode in this link.



December 1, 2023by Daniel Abib0

We had a great time attending the 2023 INFORMS Annual Meeting that took place between October 15th to October 18th in Phoenix, AZ. This is one of the largest conferences in the field of OR, with 6,000+ attendees and 1,400+ sessions.

The BAL had a strong presence at the conference with four presentations:

  • On October 15th, we had two presentations:
    • Cristián Bravo presented the work with Kamesh Korangi and Christophe Mues on “Large-Scale Portfolio Optimization using Graph Attention Networks”.

 

  • Daniel Abib presented the work with Cristián Bravo, Raffaella Calabrese, and María Óskarsdóttir on “Optimal Feature Split in Classification Models with Dependency”.

 

  • On October 17th, Sahab Zandi presented the work with Kamesh Korangi, María Óskarsdóttir, Christophe Mues, and Cristián Bravo on “Leveraging Dynamic Multilayer Networks for Modelling Credit Risk Contagion in SMEs”.

 

  • On October 18th, Yuhao (Jet) Zhou presented the work with Collins Ntim, María Óskarsdóttir, Matthew Davison, and Cristián Bravo on “Uncovering the Network Power Gap: A Deep Learning Approach to Investigating Gender Disparities in the Boardrooms of Canadian Public Firms”.

 

We also had fun trying a few good restaurants in Scottsdale, AZ!

This was a great chance to showcase the preprints that will come out in the next few months. Stay tuned for them!