DRMacIver's Notebook

You only need shallow justifications

You only need shallow justifications

This started off on the newsletter and then wandered in a direction I didn’t intend and as a result stalled at turning into something adequately newsletter shaped, so I’ve moved it over here to finish it off in whatever form it wants to be in.


I have a concept of “the legitimacy of conflict” that I’ve been trying to articulate for ages and failing to really get a good handle on. I’ve recently had a hit tweet that I think gives me part of that handle.

The tweet goes as follows:

One mistake people make in salary negotiations is that they try to make a practical or moral argument for the higher value. This can be useful, but I’ve had surprisingly good luck recently with words to the tune of “I like money. How about you offer more money than that?”

The point is that you don’t need to justify wanting more money. Obviously you want more money. All you need to communicate is that the amount of money that has currently been offered isn’t enough for you to say yes on the spot, and more would be welcomed.

Sometimes they will just say yes and offer you more money. Generally speaking, hiring is expensive and finding a good candidate is hard, and by the time you’ve got to this point, there’s a pretty wide range of amounts they’re willing to pay, and they haven’t offered you the highest amount they’d be willing off the bat.

Sometimes they will flat out say no. Sometimes they’ll even mean it. If you want more money at this point you’ll need to either walk away or break out a more serious negotiating toolkit.

Most of the time this will result in a further discussion about the value you bring, or how to justify the decision to someone else, or something of that ilk. Essentially neverOK, essentially never under the assumption that you’re operating in a relatively high strength negotiating position. There’s a decent chance of this going badly in e.g. minimum wage jobs that are oversupplied with candidates and run by petty tyrants. will they ask you why you want more money, or descend into some Oliver Twist style rant about how dare you ask for more you miserable little urchin.

Generally speaking though, in a remotely healthy environment the discussion will never involve asking why you want more money, it will always be about why you are worth more money to the company. Because, from a business point of view, of course you want more money. Money is good. Sure maybe you need it to pay the rent, or to care for your seventeen children, or one of any other reasons you might have, but this does not actually affect any of the basic business reality, which is that it’s perfectly reasonable for you to want more money and it’s also perfectly reasonable for them to not to spend more money.Now the reality is that the person you’re talking to does not have incentives perfectly aligned with the business, and as a result telling them a sob story about how you’ve got all these cute orphans you’re looking after and you need the extra money to buy them expensive medicine for their rare genetic diseases please sir won’t you think of the orphans, might well sway them to try to work a bit harder to offer you more money. But now, functionally, you’ve just got a repeat of the general situation with you and them working together to negotiate with the broader business which is, ultimately, not in the business of curing rare genetic diseases in orphans for free, but instead makes widgets.

A negotiation like this is a great example of what I mean when I talk about a conflict, and why conflict is legitimate.

Specifically, I’m using “conflict” in the economics sense of Thomas Schelling, as articulated in his great book “The Strategy of Conflict”: A conflict is a positive-sum game with imperfect alignment.

That is, you have a situation with two (or more) parties in which:

There are outcomes which are better for all parties than not participating in the situation.

Among those outcomes, there is disagreement as to which is the best outcome.

Whenever I use the word “conflict” in this article, I mean it in this sense of conflicting desires in a situation where it is to your mutual benefit to figure out an outcome compatible with them both.

A salary negotiation is precisely this sort of scenario: Participation is getting hired (which requires both your agreement), and getting hired is at this point in the process presumed to be better for both of you than not getting hired (you want the job, they want someone to do the job and think you would be a good such someone), but the best outcome for you is that they pay you a lot, and the best outcome for them is that they pay you a little.There are all sorts of complexities to this about retention, ethics, etc. Try to ignore those. At any rate, even to the degree that those matter, they matter mostly for large changes. A 1% difference either way probably doesn’t affect either of those that much, but a 1% pay raise is clearly slightly better for you, and a 1% pay decrease is clearly slightly better for them.

One of the difficulties I’ve had in articulating the legitimacy of conflict is that people hear “conflict” and they think “fight”, where Schelling and I mean it more in the sense of “things that are in tension with each other that you need to resolve to make the situation work well”.

The other difficulty is that part of the problem I’m pointing at is that I don’t think people emotionally distinguish the two and do, in fact, experience something like this as if it were a fight.

No correct answers

I think one reason people experience the salary negotiation as a fight is that it feels too easy to frame it in moral terms. There’s an amount of money you “deserve” and if you ask for more than that you’re greedy and if you get less than that the company is greedy. Either way someone is bad. This viewpoint is, I must emphasise, wrong, and you’ll make less miserable and earn less money than you could if you adopt it, but it’s a very easy trap to fall into. But I think it’s hard to see that starting from this scenario.

Here’s another example, I think originally from Schelling:Though his originally example was probably a bit different. I think it involved the movies? Suppose you and I want to hang out, and settle on going out for a meal. We rule out anything literally unacceptable to one of us (e.g. as much as I’d like to I can’t eat pizza, you need options that aren’t super spicy) and settle on a number of options that we’d be fine with. My favourite is the Thai option, yours is the Mexican option. Either of us would be fine with either option, but each of us has a clear preference for one over the other.Why not find a third option that we both like? Well, what are you, some sort of fancy big city type with lots of restaurant options? In my small town, you’re lucky if there’s one restaurant you like! Two is just luxury.

More seriously, this is always a tempting rebuttal but I think basically doesn’t work. There just sometimes aren’t such options. Some people really do have incompatible food preferences where you’re never going to find a meal that delights them both, and “delights one of them and is more or less fine for the other” is probably the best outcome you can achieve.

This is particularly true once you add hard constraints. Try feeding a vegan at the same time as someone with soy and nut allergies and see how well that goes for you.

This is a clear conflict: We both want to go out for a meal. We don’t agree which restaurant is best, but we’d prefer either to not going out for a meal.

It would also, I feel, be utterly idiotic for this conflict to turn into a fight. We’re friends. It’s not just that we can resolve this amicably, it’s that it getting unamicable in the first place at all would be a little weird. We can flip a coin, or take turns deciding, or you can decide where we eat and I decide what bar we go to after, or adopt one of a variety of other resolution mechanisms, and it will be fine.

One thing we shouldn’t do is decide on the basis of who wants their outcome more. Or, we might do that sometimes - if there’s a cool new Mexican place in town that you’ve been really excited to try out, and I want to go to the Thai place I’m familiar with because it’s consistently pretty good, it seems reasonable that we should pick the Mexican place. If, however, every time we go out you’re like this, and I never really acquire your taste for novelty despite our best efforts, I think I’d quite reasonably feel a little hard done by if my preferences never got a hearing.This may seem like a suspiciously specific example, but it’s honestly entirely fictional.

Because this is a negotiation as part of an ongoing relationship, it’s clear that there doesn’t have to be a single right answer to the conflict in desires, you resolve so that mostly everybody is happy. I originally wrote “so that everything averages out over time” but actually I think this is not correct. It may well just be the case that the result is consistently “unfair” but still better than not participating. Many different dynamics can lead to this, and it’s not intrinsically a problem.

Resource Constraints

Another example of a conflict I often use is: Do you take the last Eclair in the shop? Someone else might want it after all!This is a slightly strange example of a positive sum game because although participating is never worse than not participating, unless there’s an Eclair there it’s not really better either. Unless you needed to go to the shop anyway.

I think, clearly, the answer is yes.Although when I’m buying a lot of something, I will usually make sure to leave at least one on the shelves. e.g. there’s a particular frozen chicken curry that the child is currently obsessed with and that we’re perfectly happy to be a default meal for her, so I bought a bunch of them recently, and made sure to leave one in the freezer. It’ll get restocked soon, but in the immediate future if there’s someone looking for a mediocre Morrisons own brand butter chicken, they won’t go lacking.

Abandonment note

At this point I stopped writing and I’ve lost enough context that I no longer intend to resume at this point. I did say that it’s hard for me to write about this subject!