Xinja

» see xinja.com.au as it lived, on the wayback machine

~ the goodbye ~

xinja was australia-only at the regulatory layer. APRA-supervised, deposits held in AUD. nothing about the product could touch a customer in ankara because the license didn't reach here. i am writing this from techmeme distance, computer engineer in turkey reading about neobank collapses the same way one reads about foreign-market bankruptcies, at a delay.

eric wilson founded xinja in sydney in 2017. it was one of the first australian challenger banks. 86 400 and judo bank were in the same cohort, and volt came in shortly after. all four tried to do what monzo and n26 had done in europe. app-first, aiming to beat the big four on savings rate. wilson raised through crowdfunding twice and hit the retail investor limit both times. APRA granted a restricted banking license in december 2018 and upgraded it to a full ADI on 9 september 2019, making xinja the second neobank in the country to get one.

the transaction and savings accounts launched on 15 january 2020. the flagship product was the stash saver, an on-demand savings account paying 2.25 percent when the RBA cash rate was 0.75 percent. the rate was set higher than xinja could actually earn on customer deposits, but the plan was to raise capital fast enough to bridge the gap. deposits arrived faster than xinja expected. inside two months the platform had roughly $500 million in customer money.

~ the chronicle ~

~ the breakout ~

$500 million in deposits with no lending book meant xinja was paying deposit interest without earning lending interest to cover it. the plan had always been to launch a lending product to close the gap. lending never launched. wilson had a term sheet in march 2020 with a dubai-based investment group called world investments for A$433 million, and that was the round that would fund the lending build-out plus the deposit cost drag.

covid landed in march 2020. dubai-based capital pulled back on international commitments almost immediately as the world's markets went sideways. wilson kept the term sheet alive through the northern spring but the closing conditions kept slipping. by mid-year the stash rate had been quietly cut to 1.8 percent, then paused on new deposits entirely. the on-demand mechanic that had brought customers in was being unwound from underneath them.

~ what really happened ~

xinja announced on 15 december 2020 that it would return its ADI license to APRA. all customer deposits were refunded within a few weeks, no loss to individual account holders, which was one of the key differences from the wirecard collapse six months earlier. the fintech shell continued for a stretch on the money-management app that had preceded the bank, though that faded through 2021.

the failure mode was mostly identifiable in real time. a neobank in a small low-margin market needed scale before it needed lending. the covid capital freeze hit before the lending arm was ready and before deposit growth could be paced against operational cost. no fraud and no scandal, just a runway that dried up during the specific quarter that dubai capital was not deployable. 86 400 got acquired by NAB in 2021 and judo bank IPO'd. volt itself handed its license back in june 2022. xinja was the first of that cohort to go.

~ what replaced it ~

xinja is a study in what happens when a neobank hits the deposit side hard before the lending side is built. every neobank textbook from 2021 forward cites the stash rate as the specific mistake, because it drew in the wrong kind of customer at the wrong pace against a balance sheet that could not yet earn the rate back. simple in the US, in a different way, hit the wall two years later with the same underlying problem, though it had bbva absorbing the loss instead of a public license handback.

the whole australian neobank experiment of 2017 to 2022 mostly did not survive as an independent cohort. only judo, which focused on business lending from day one and never chased consumer savings, made it through as a going concern. xinja's remaining assets and its lending-team hires were absorbed into that broader recovery, though not under the xinja name.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »