Cellphone data study: Reynosa shoppers drove roughly $48 million in McAllen retail sales across 33 peak days, economist says

McAllen hosted a presentation on cross‑border commerce where an economist described a cellphone‑data method to estimate spending by shoppers who cross from Reynosa, Mexico, into McAllen.

Dr. Belinda Roman, introduced as an associate professor of economics at Saint Mary's University, said researchers used anonymized cellular data that originated in Reynosa, isolated residential devices (required to be in a Reynosa neighborhood for 48 hours before crossing) and matched device locations to Mexican household statistics to estimate spending patterns. "The Mexican shopper has money to spend," Roman said.

Roman summarized the data and conservative assumptions used to translate device movements into spending figures: the dataset covered selected peak shopping windows (the week before and after Christmas, Semana Santa, and the Black Friday period), approximately 33 days in total; researchers observed roughly 140,000 unique devices during those sample periods, visiting more than 800 McAllen businesses. The analysis found about 177,000,000 visits to McAllen locations across those 33 days, an average dwell time of about 20–25 minutes per visit and a median household income for the crossing shoppers of about $45,000–$46,000 (as measured using Mexican statistics matched to device origin).

Using a conservative per‑person spending estimate of $113, Roman reported period estimates of roughly $15 million (Christmas window), $17 million (Holy Week) and $16 million (Black Friday), producing about $48 million in direct spending over the sampled 33 days. When the team expanded the calculation to include indirect and induced effects, Roman said the estimated economic impact for McAllen totaled approximately $80 million for the 33‑day sample.

Roman described specific methodological choices: she excluded transient commercial traffic and treated devices spending more than eight hours in McAllen as likely workers rather than shoppers; she noted that most observed transactions were cash withdrawals at ATMs and that using only credit‑card data would miss a substantial portion of activity. She said the dataset began with nearly one billion raw data points that were reduced and cleaned to isolate residential shopper behavior.

Roman recommended several follow‑up actions for city economic development: a retail program targeted at middle‑income cross‑border shoppers, marketing that signals welcome to varying shopper incomes, leveraging ATM and cash‑usage patterns in communications and business outreach, and exploring technology and financial‑infrastructure linkages that support cross‑border commerce.

The presentation did not attach an ordinance, budget appropriation or formal city action; Roman said she would share "as much information as I can about what we have available" when asked about data availability.