Climbing the Pareto Curve

Pareto - the name has been coming up again and again. First a trickle, then suddenly a wall of water as high and steep as a tsunami wave. How quickly you realize that you're faced with a seemingly unsurmountable challenge.

Perhaps it's just frequency illusion. As a fellow science academic in training, we're meant to push the boundaries. Every once in a while, or recently even, I've been faced with just this problem of doing something that's not quite been done before, and having to come up with an intricate solution to tackle it.

The easy part is just in repurposing old code from my past projects to do a slightly different thing. The hard part is when you face fantastic stuff that is wonderfully engineered. To the point that it is designed to be 'easy' and very opinionated on how to keep it that way. Think Apple-friendly, great for the masses, not so great for the tinkerers amongst us.

The topic of 20% pain for 80% gain (and vice versa) has actually been on my to-write list for a while. It's an interesting concept, on how 'a little goes a long way'. You stroll along casually in life, doing what you do, and suddenly to improve/upgrade/keep-up, you realize how much effort it takes to get there, and stay there.

The thing you need to grasp too, is that the Pareto principle is not an absolute thing, but a relative one. In other words, it's about ranking. Yes, it is when you're having to compare yourself to others or keep up with trends, that you'll be faced with this problem.

Now in a Zen/Chan-ful life, you wouldn't even have to care about these petty differences. But let's for the sake of argument, presume that you want to improve, you want to be richer, you want to be better.

Just about everyone, the 80%, will be putting in that 20% effort required to maintain their average position in the masses. You might be contempt that you're much better than before, and it's okay to have that attitude.

I'll use my mom as an example. She told me once, that a few of her friends couldn't believe looking at what she's like now, that she had a pretty 'hard' childhood. By hard, that means having to get water from a well, not from a pipe.

For those of you who think that water comes from pipes and money comes out of walls (ATMs), that means throwing a bucket into a deep hole (with a certain technique) and pulling it up several metres on a pulley. Repeat that for a family of ten - every day.

That was maybe forty years ago now. These few years, she's gotten in touch with a lot of her old school friends from back in the day. I've met a few of them on Chinese New Year trips back, and you can see none of them living in old single storey wooden houses. Heck, there's actually a few living in condos - two storey ones - with an amazing night view!

So that's what I mean by relative. Over the years, it's not too hard to put in that 20% effort to keep up with technological improvements. Things like piped water we take for granted now, electricity to charge our mobile phones every day, or having a concrete/steel/fibreglass air-conditioned/heated home not subject to termites.

But if you want to be in that top 20% of the population, well, welcome to the 80% pain game.

Let's take university degrees for example. Back in my mom's time, a diploma would have been a feat in itself. To be fair, even finishing secondary school would have been a good thing.

Now? You'll probably graduate with a university degree just like 80% of your peers. Sure there are some who will go to do a trade, or start their own business or something. But if you're like me with Asian parents, you'll probably have to go to university.

Having finished my Honours postgraduate degree back in 2015 (something like a two years Masters condensed into one year), I knew I was pretty much the same with most of my peers. In other words, I had not moved much on the Pareto curve. Not that I care much about my ranking actually, I'm just stating a fact.

Once I got out on the workforce, it really struck me how average having a degree actually is. So average, that I do think the 'no-degree required' movement has its merits.

I knew I was more a study/research-type person by heart actually. One of my old colleagues who's now in Australia used to joke that I'm more of an academic. A long while back, I had a random chat with an old friend (now in an Ivy League uni?) after I graduated, and she was probably the only one I knew who was also considering the academic route back then. We talked about grad school, tenure, and all that, whereas any other one of my friends would be on the job seeking side.

Fast forward to 2018, it's almost like Masters are the new Bachelor's degree. I imagine my PhD will not be looked as anything special by the time I graduate. Again, just to reiterate, the Pareto Curve doesn't really change, it's just that our baseline expectation goes up.

It's like one of Zeno's paradoxes, you feel like you're catching up to the tortoise in front of you, but in reality, you're still second place. Or that trick question: "In a race, what is your position if you overtake the 3rd fastest person?" If you thought 2nd, wrong, you're just the 3rd.

That 80% of the effort goes into climbing the very steep top 20% of the curve, is really just insane!! You're so much better off doing the little things that will take you 80% of the way there.

But the funny thing about competitive humans is, we don't just settle for that easy 80% now do we? In my previous workplace, you could easily see the 80/20 principle at work. A good 20% of our time was spent on recurring stuff that easily covers 80% of our revenue. Yet we spend 80% of our time on chasing new leads, handling demanding customers, designing custom tools, just for that measly 20%.

You could almost argue that the extra 20% isn't worth the effort. Why employ an army of people to deal with the tricky clients, when cutting those clients loose won't cause that big a dent anyway? Maybe it's the hope that the hard 20% might one day become part of the easy 80%.

Put it this way, say you work five days a week. Four out of five days, or 80% of the time isn't really spent doing anything much right, let's be honest. You could squeeze everything into one day, or just 20% of your work week, and get most of your deliverables sorted. Imagine that, having to only commute 1 day a week to work!

Sadly, it sounds good in theory, but it practically doesn't work that way. First off, you'll need to convince your superiors for that level of work flexibility. Second, you can't quite schedule all your meetings and emails to fall on one day. Third, the 80% of the time not doing anything productive is still essential for your mental health and creativity.

I've taken it to the extreme once, when I was transitioning into my PhD, doing part-time work and part-time studies. It seemed logical that I could spend 20 hours instead of 40 hours producing the same output in my workplace, and that was approximately true because I automated a lot of stuff. For my PhD though, which required a good degree of creativity at the start, I was spending maybe 30 hours, eating into Saturday, to get some solid output. Plus some nights I worked as well, so it was a good 50-60 hours if I'm honest with you.

There's a reason we sleep, because even though it seems like a waste of time, that's when our body and mind can recover from our daytime activities. Yes, I don't regret the 50-60 hour work week I went through, and I wouldn't hesitate to do it again, but it goes to show what climbing the Pareto curve entails.

On one hand, I was earning a good amount of money, enough to cover not only my living costs but also my sis's ~NZD30,000 university tuition fees this year + living costs. On the other hand, the 80% effort being put in to get that 20% extra, is costly. I hardly had time to socialize (not that I minded), used nights to study (watching video lectures/tutorials online) and had to pay more tax since I ended up at a higher tax bracket!

So the big question is, is it worth it to you?

Reader you reading this might be thinking: "Just tell me whether I should be putting in that 80% extra or not!" Well, like a good academic, the answer really is: "It depends."

Working that 50-60 hour work week allowed me to save up a lot, and by being frugal and investing that lot, it gave me a good head start on my road to financial independence (FI). It's sacrificing eating the marshmallow now so that I can have more in the future. Heck, I calculated that I had a ~80% savings rate at one point!

If you're happy where you are, I think that's a good thing too. To be fair, if you're smart enough, you can put in that tiny 20% of effort, say 20% of your salary, and you can be FI in less than 30 years. Put in another 20% to save 40%, and you shave 10 years off, becoming FI in 20 years. Yet another 20%? With a savings rate of 60%, you'll FI in about 10 years!

From this blog post I keep linking to, here's how it looks:

Notice the power law at play. Once you start saving more than 70-80% of your income, you're not cutting your FI date by very much. You could be clipping coupons, cooking in bulk, dumpster diving and more. All that effort! For what meager gain?

The Pareto curve - something hard to climb up, but easy to slide down.

Something less discussed is Pareto efficiency - where "no one can be made better off without some being made worse off" If I may paraphrase, it is akin to the optimal place on the Pareto curve, the ideal place on the pain/gain spectrum.

The thing I'm working on for my PhD right now? I have an idea for the model I alluded to in this previous optimizing post. To use this thing called a GAN wherein two neural networks are competing in a Zero-sum game that (hopefully) results in a Pareto optimal state. This blog post tries to explain it.

Conceptually, it's an interesting way to think. You push something too far, it becomes too painful, so you reverse it. You don't push it far enough, it's not getting a good enough result, so you push harder. I like to think of windows here in New Zealand as an air-conditioner. Open it too wide, and it gets a bit cold. Close it too much, and you don't get much air flow.

Of course, it's much more complicated than that. Instead of one button to push though, it might be many little knobs and levers to control. Most of them make sense because they're labelled, it's the unlabelled ones that trip me up. But that's a problem for me to solve by the end of the month when September ends.

So yeah, where do you want to settle at? Do you want to challenge the steep Pareto Curve? Or are you happy where you are?

There is a sweet spot, go find it.