After graduating from NUS, I returned to the company I co-founded to lead a small team called the ‘Acceleration Task Force’, which combines corporate strategy and business transformation. One of my early tasks here is building the Human Resource Department from the ground up, as it has been three years since we last had one (why and how we operated a 100-employee company without HRD is a story for another post). Part of starting the HRD is supervising and mentoring the newly appointed Head of HR, and in the process, I was reminded of recruitment challenges we always have as a company headquartered in Kediri, a small city in East Java, Indonesia.
The Pattern That Keeps Appearing
It is challenging to find good talent, and when we see them somewhere else, they don’t want to move to Kediri. Last month, we opened a job vacancy for a TikTok Content Creator position. There were more than 30 applicants, but only a few have content creation or social media experience. Long story short, we selected a candidate from Malang, a neighbouring city, but she said she preferred another offer in her hometown. Then, we offered the position to two alumni who have good TikTok presence, but both turned down the offer as they do not like the idea of working in Kediri. We later scouted three candidates on TikTok and adjusted the work-from-office requirement to just a week per month: a candidate expected a very high salary we could not afford, another one already had a full-time job, and another one was also reluctant to come to Kediri once a month.
That is not a unique story. We found a similar problem when we posted a vacancy for a graphic designer position a few years ago. There are many skilled applicants in terms of the ability to use design tools or applications, but only a few have good design taste at the level of companies in big cities. Another problem appeared when we opened vacancies for Offline TOEFL and IELTS teachers. All the applicants are from outside Kediri, and since the minimum wage in their hometowns is higher than in Kediri, they expect a higher salary, even though the position was intended for an on-site job in Kediri. Similarly, candidates for our Head of Online Programs are also reluctant to move to Kediri even though the salary offer met their expectations.
What’s Actually Happening
There are multiple forces at work. The first is what economists call a thin labour market: there are not many workers and not many employers competing for them, so matching becomes inefficient. But the problem goes beyond quantity. Even when applicants exist, their skills often do not align with what we need. The graphic designer case illustrates this clearly: we received many applications from people who can operate Adobe Photoshop, Illustrator, and Canva, but operating tools and having design taste are two different things. Taste is developed through exposure, working alongside better designers, receiving feedback, and seeing high standards daily. This is what economists call agglomeration effects: skills develop faster in dense markets because learning becomes ambient. In Kediri, that ecosystem barely exists. So the market is thin, and what remains is often mismatched.
The second force is spatial sorting, or what people commonly call brain drain. Ambitious workers leave for cities with more opportunities, and this creates a self-reinforcing cycle. When we found qualified candidates in Malang or through TikTok scouting, they did not want to come, not necessarily because of money, but because Kediri is not where they want to build their careers. The opportunity cost of living here, fewer job options, thinner professional networks, and less career flexibility, outweigh what we can offer. The TOEFL and IELTS teacher vacancies show a related but different problem: spatial wage differentials blocking the match entirely. Candidates from other cities have reservation wages, the minimum they will accept, anchored to their hometown’s cost of living and minimum wage standards. Kediri’s wage structure simply cannot meet those expectations, even though the candidates are willing to apply and the jobs are available. The workers exist, the vacancies exist, but the deal does not close. Meanwhile, the candidates who stay in Kediri are often those without better alternatives.
The Hidden Cost
Since our solution is choosing the best among those who are available, we often have to train them from zero. In the offline teachers’ case, we developed a scholarship scheme where students can get free four to five-month English courses, a TOEFL or IELTS test, and teacher trainings, in exchange for a 1.5 to two-year teaching contract with our standard salary. For TikTok content creators, we developed a very detailed tutorial, down to what words to say/write in the hook and closing, on how to create a viral content, hold a weekly evaluation and mentoring with the social media manager, and create 3-month targets that increase gradually starting from very low target with the assumption that the talent starts learning from the basics.
Those decisions mean we pay in at least three ways: in wages, in building what the market didn’t provide, and for slower output. For the offline teachers’ development, we spent around Kediri’s minimum wage per month (IDR 2,5 million) per person for a minimum of five months before they are ready to teach. We also often send our new employees to training that is worth more than that. Due to this cost consideration, we often deliberately choose candidates with lower qualifications and a lower reservation wage to compensate for training expenditures. For the TikTok content creator, by employing a newbie, we lose around 35 sales per month (worth approximately IDR 40 million) that we normally receive when we have a skilled talent. If we calculate managerial attention, the cost we spend would even get bigger.
What It Means for Regional Development
Indonesia talks constantly about decentralization and developing regions beyond Jakarta. The government wants economic activity to spread, wants secondary and smaller cities to grow, and wants to reduce the gravitational pull of Java’s major metros. But talent ecosystems do not decentralize automatically. If companies in smaller cities cannot access skilled workers, they either stay small, operate at lower quality, or relocate to where the talent already is. This limits job creation in places like Kediri, which ironically reinforces the migration pressure toward big cities.
Our experience suggests that the cost of doing business in smaller cities is not as low as it appears. Yes, minimum wages are lower, and rent is cheaper. But when you account for training investments, slower productivity during learning periods, and the managerial attention required to build capabilities from scratch, the gap narrows significantly. In some cases, it might even reverse. Policymakers who focus only on nominal wage differences miss this hidden subsidy that employers in smaller cities provide. We are essentially doing workforce development that the education system and the market did not deliver.
I do not have a complete answer for what would fix this. Better local universities with stronger industry links might help. Incentives for skilled workers to return to their hometowns could work, though I am skeptical. What I do know is that building a company in Kediri means accepting that we are a training institution whether we planned to be or not.