Hacker News Reader

Comment by alistairSH | Hacker News Reader

if they do train the person, then that person will just then go walk off with the training experience and work at another firm.

That's the free market at work. If a trained [skilled job] is worth $100k, but you're only willing to pay $90k because "they're new", they should walk across the street.

I absolutely hate when self-declared "free market" believers don't understand that there are two sides to every market. And employment is just another market. (not claiming you're making that mistake, but it's one I see frequently when wage discussions come up)

View on HN

I think you're missing the deeper point.

In a highly mobile labour market training people makes no sense. A trained workforce is a commons, and what we see is a classic tragedy of the commons.

Consider two firms with a budget of $100,000 for an employee for a year. Firm A decides to offer wages of $50k to an untrained worker and invest the other $50k in an intensive 3 month training program. Firm B decides to do no training at all. Worker joins firm A, benefits from the three months training and then immediately quits to go pick up the $100k salary for the rest of the year.

Generic job training is an investment, but, there's no way to get reliable ROI on it unless the training is super firm specific (like training in your own trade secrets). So nobody trains and everything is pushed onto the employees themselves via universities and trade schools, because that resolves the tragedy of the commons.

They could just raise wages once they've upskilled one of their employees so that they don't leave. Problem is that internal raises beyond inflation are basically impossible for most workers, so the only thing that makes sense is to leave a job 1-2 years in for a double digit raise.

It seems like this can easily be fixed by having people sign up for contracts that last more than 1 year or at least have some financial incentives that accrue to employees who stay longer? For example, tech companies have stocks that only vest after 4 years. I don't see why in principle this couldn't be even longer.

I don't know much about it personally, but I think the US government has a lot of benefits (like retirement) that scales with years of service.

I actually suspect that at least for many big companies the problem isn't getting people to stay after training, it's getting the best employees to stay and getting lower-performing employees to leave. I think for this reason, most tech companies structure pay with a 4-year contract, which promises a high salary for 4-years (due to bonuses and stock), and then an implicit pay drop if there's no promotion after 4-years.

Having seen this up close and being a good friend to both the owners and the newbs, the market based approach is non-optimal in this stiuation.

People want to be hired and trained, and need to be trained to preform the work. The issue, especially at smaller firms, is that training takes a lot of resources that aren't easy to spare. As such, the year costs and benefits aren't nearly as close.

As an extreme and simplified example: The new hire needs to be paid $50k/year. Also, the profit off their labor is $10k/year (firm makes $60k, pays employee $50k). However, training lasts 6 months. So, for half the year, they are not net profitable. That means that they are net negative their first year (0.5 X $60k = $30k lost in 6 months, add that to $50k paid out and you get $80k as the total 'cost' of the newbie at year's end. You end up at -$20k for the year for the newbie). So, to make back the cost of training this newbie, you would need for that employee to work for you for two more years (2 years X $10k/year profit = $20k. The training cost was -$20k). Then you would be net neutral. Also, not only is the new hire not 'pulling their weight', but they also take time and energy away from other employees.

To add to this, the newbie can just walk out and get hired somewhere else at any time. If that were to occur, then you really have to eat the costs of the traning and will not see any investment come back. In fact, it is likely to be going to help out a competitor.

Now, the above is an extreme example, of course. But some form of this calculus is occuring at smaller firms. Perhaps it's only six weeks of training that occurs. Perhaps they make much more profit off of the employees. Perhaps your business is growing fast and your are getting more efficient. Still, there is a significant cost to training newbies, one that many firms are loathe to pay out.

Now, economists have been talking about these issues for a LONG time. This kind of 'capital' investment mechanism is well studied, and , yes, a market is a very good way of sussing these things out.

However, we also know that markets are rational, and humans in a market are far from rational. These 'sunk cost' fallacies are very painful to humans. I suspect that they are very painful when it comes to training. You work with a person for weeks to months, watch them grow and learn, get to know them well, you put time and effort into them and come to care about them. Then the market comes in and takes all that emotional investment away, because 'prices'. It's not easy and it is not rational, it becomes (in some degree) personal.

So, yes, the employment market is a just like any other market. But training is more than that to us humans. That kind of investment is not something that can always be priced.

Reply on news.ycombinator.com