The Request Batting Average

There may or may not be stupid questions, but there are definitely stupid requests. I have a concept of a “request batting average,” computed as [number of approvals] divided by [number of requests]. You can tell a lot about the quality of an engineer asking for money, or a Christian saying his prayers, by looking at the relevant batting average.

Opportunities to spend money constantly confront a refinery engineer. This valve doesn’t shut very tightly; we could get a new one. We could repair this heat exchanger less frequently if we bought one made of stainless steel. This sixty-year-old pump is not designed to modern standards, and its mean-time-between-repairs suffers for it. That pipe could use a paint job. Maybe we should tear down that pressure vessel that has been sitting there unused for longer than I’ve been alive. This refinery would be easier to manage if we just built a new one made entirely from indestructible nickel-moly-chrome alloys right beside it, and then abandoned the old one. All of these ideas have some degree of aesthetic appeal, would be desirable if free, and would in some sense result in a “better” refinery.

But the plant manager cannot possibly fund every pleasant engineering idea. He operates with a basically fixed budget, so ideas compete for funds. And even if there is slack in the budget, the plant manager also wants the refinery to achieve its broader purposes, notably making net cash so the company can pay the dividend to its owners, the Benevolent Ladies’ Association Retirement Fund (BLARF, for short) and other stockholders.

An engineer could just say “Well, I will just throw out as many nice ideas as I can, and management can pick which ones they want!” Then management has to wade through a flood of half-baked proposals, most of which are not aligned with their priorities. The pattern will quickly become clear, as will a corresponding pattern where the engineer’s requests meet with skepticism. His batting average will be low. And since only 20% of his past ideas got funded, his future ideas are assumed to have only a 20% chance of meriting funding.

Alternately, the engineer can triage ideas himself. He must develop a sense of what management is likely to approve and critically evaluate, in light of this, whether a given idea is worth developing further. Each idea he chooses to develop should become completed staff work, a fully developed concept and justification such that predictable questions are answered and difficulties resolved before presentation to management. This engineer will have a high batting average, and managers will quickly recognize that when he asks for money, they are usually going to say yes.

Now, an engineer should be willing to pursue non-obvious ideas, rather than just parroting management. Sometimes it takes significant development to see whether an idea is good, and it’s often worth doing the work even when you know there’s a 75% chance that you yourself will later kill the idea. And it would be a mistake to target a 100% batting average; it’s hard to be perfectly calibrated on what management will want, and management does want to see options and make choices. But an 80% batting average may be achievable and desirable—so if you have twenty ideas and you expect management can only fund four, give them the top five. Maybe your #5 idea is better than you think, and you’re missing key information that disqualifies your #3. (And in any case, managers need to say “no” occasionally just so everyone can feel like they are doing their jobs, themselves included.)

A good engineer will focus efforts and requests on things management is likely to be interested in. You want this kind of engineer working for you. You want to promote this kind of engineer to management, given that they are demonstrating an ability to do management prioritization and drive towards the organization’s goals.

There are parallels between well-done funding requests and effective prayer. So think about a prayer batting average, [number of approvals] over [number of requests].

You could pray and ask for any number of things. Please bless me with perpetual health and beauty. Please bless me to know if a parenting idea I’ve been thinking through is on the right track. Please bless this double-chocolate brownie to “strengthen and nourish my body,” by some miraculous violation of all known nutritional science. Please guide me with thy Spirit in a situation I’m about to face. Please give me specific needed blessings me as I work to fulfill this worthy goal. Please bless that I can get lots of Lego for Christmas, and also win the lottery. Please bless me with spiritual strength to “submit to all things which the Lord seeth fit to inflict upon [me], even as a child doth submit to his father” (Mosiah 3:19).

God isn’t constrained by a budget; His interest in the pensioners of BLARF goes beyond the fiduciary. But He is still constrained – constrained by His goals. He does not wish to eliminate free will, remove all of life’s challenges, or produce spectacular miracles on demand. He’s been pretty clear about this, too.

Thus I don’t think God is particularly impressed by desire-vomit prayers. You can look in the mirror and pretty well intuit that perpetual beauty is not a part of God’s plan for you (except for you, wife). If lottery wealth was actually going to be good for you in the eternal long run, you’d think God would have created the world to have more of it. And so forth ad nauseam. Asking for this sort of thing results in a low prayer batting average. It could result in frustration with prayer.

The scriptures gently point us away from simply throwing out low-quality requests. Thus in Doctrine and Covenants 9:7-8 the Lord said “Behold, you have not understood; you have supposed that I would give it unto you, when you took no thought save it was to ask me. But, behold, I say unto you, that you must study it out in your mind; then you must ask me if it be right…”

Spiritually mature prayers will involve well-considered requests, and the supplicant should have a reasonable expectation that the request will be granted. Getting to know the approver and his plans, and triaging requests accordingly, is a worthwhile effort. Nephi (the son of Helaman) was outstanding at this, to the point that Christ said to him “I will make thee mighty in word and in deed, in faith and in works; yea, even that all things shall be done unto thee according to thy word, for thou shalt not ask that which is contrary to my will” (Helaman 10:5). This is like blank-check spending authority—I approve everything you ask for anyways, so you can just go ahead.

Similarly, he told the Twelve that “whatsoever ye shall ask in my name, that will I do, that the Father may be glorified in the Son” (John 14:13). Speaking to a general audience, Christ made a slightly different promise: “And whatsoever ye shall ask the Father in my name, which is right, believing that ye shall receive, behold it shall be given unto you” (3 Nephi 18:20). So we need to make sure we are asking for that “which is right,” and for things that we believe we “shall receive.”

Like in the engineering case, it is sometimes hard to know beforehand exactly which requests will align with the approver’s goals, so we shouldn’t be excessively cautious about bringing up ideas. God wants to know what we are going through, prayer should have a conversational element, and we should not be too fearful – as Paul teaches “Let us therefore come boldly unto the throne of grace, that we may obtain mercy, and find grace to help in time of need” (Hebrews 4:16). And we have Christ’s example of praying “O my Father, if it be possible, let this cup pass from me: nevertheless not as I will, but as thou wilt” (Matthew 26:39). This shows both a willingness to be open with God and make a request that may not be approved, and a recognition and acceptance that it likely will not be.

We can and should let our prayers benefit from experience, make sure we are thinking them through, and target a relatively high batting average.

Don’t ask for stupid stuff!

My Stock Advice for Engineering Students

Good engineering students often do a poor job of managing their career opportunities. I want to help.

I’ve participated in career fairs and job interviews various times on each side of the table, hiring and searching. I have been in the room making decisions on who to interview. I’ve been in the room making decisions on who to hire. I’ve offered people internships. I’ve managed interns, and recommended them (or not) as permanent hires. So I have some understanding of how all of this works, and a whole series of stock rants about it.

Often, I interview awesome people who probably have very high potential, but have not optimally developed that potential, or haven’t done what is necessary to communicate their potential clearly. I want engineers to do better—especially BYU engineers. Thus I share my compiled rant/wisdom, such as it is.

(This should go without saying, but this is on my personal site, and it reflects my personal opinions—I do not speak for my employer, Valero.)

Goal: a traditional engineering job

There are various career paths open to an engineer. In this post I’m assuming your goal is to get a traditional engineering job after graduating with a bachelor’s degree. I have an oil & gas perspective, but the principles seem to be similar for engineering jobs in aerospace, automotive, etc. Some engineers instead want to do consulting, patent law, academic research, etc., in which case some principles from this post may apply, but you also ought to find more specialized advice.

If you want a job after graduation, you need to try to get an internship before graduation—ideally several internships, one every summer for the length of your degree. The best way to get an internship is to have had a previous internship. So the perennial question is: how can I get my first real engineering internship? In this post I lay out advice for getting there. Then for future internships and jobs, the same general principles apply.

Focus on developing your resume

You need an attractive resume to get a good engineering job. There are many components to having an attractive resume; it’s nice if you can hit “all of the above” and be a trilingual champion athlete with a stratospheric GPA, a demonstrated commitment to community service, and prior internships in your field. Employers will talk about valuing various thing, and they even do. And you should not base your whole life on employers’ opinions.

But cut through all of that, and you really, really need to keep adequate focus on two key things:

  • Maintain a good GPA
  • Get engineering work experience

It is going to be awfully tough to get a desirable internship if you don’t bring a good GPA and some kind of engineering experience to the table. So prioritize these things over other work!

Engineering students are typically hard workers, and often have non-engineering employable skills. In the short run, they can often make more in other opportunities (random hourly jobs, skilled trade work, running a small business, pest control, whatever) than they can doing anything engineering-related. Engineering homework has a wage of zero. Doing engineering in a club has a wage of zero. Starting out in an engineering research lab often has a wage of zero.

Nonetheless, you may well be better off in the long run doing engineering work for free and focusing on holding/raising your GPA (and taking on student debt as needed), instead of working even a $25+/hr non-engineering job. The difference between getting a top engineering job opportunity and an average opportunity is potentially tens of thousands of dollars a year for your entire career, so play the long game. If you are smart enough to hack it as an engineer, you’re probably better off the sooner you commit.

(Aside: I never actually found out the hourly rate for my first engineering job. Dr. Thomas Knotts, who is great, hired me saying something like “I will pay you whatever the department standard rate is” and I said “ok” and got to work, in the full knowledge that the real pay was the knowledge I was gaining and the experience on my resume. And without telling my whole life story, skills and stories from that job directly helped me get all my other jobs I’ve had since.)

Maintain a strong GPA

As far as GPA is concerned, students often ask “what is a good GPA?” Depends on the company and the individual recruiter. I am familiar with at least two companies that have flirted with a 3.5 cutoff. In my mind, anything in the 3.8-3.9 range is great, and anything above 3.5 is good. Holding a near-perfect GPA can smack of perfectionism, and may not be optimal vs. slacking off a bit more in school and reallocating your efforts to something else (like gaining non-classroom experience). The 3.2 to 3.5 range is OK; it’s not setting you apart from the crowd, but great experience and other factors may be able to counterbalance that. Anything under 3.0 is getting into resume-killer territory. Normally sophomore-level engineers have higher GPAs than junior-level engineers; a 3.8 is a lot less impressive when few of the constituent grades came from hard engineering classes.

It is good to take hard classes early, so that you can take some easy classes later too. Do enough practice problems, spend time on the homework, etc. Avoid overextending yourself.

So enough about academics. Let’s talk about the other pillar of your resume: getting some experience.

Get a real engineering job

It can be challenging to get your foot in the door with engineering experiences, and as mentioned, it can feel like you need to have experience to get experience. My standard advice is to make a list of all the professors in your department, in order of how interesting their research is (use Google, department website, etc.). Take the professor whose research you like the best, read the last couple of academic papers he has published (Google again), and then go talk to him. “Professor X, I am [name]. Here’s my resume. I’ve read your last couple of papers on [whatever] and I think you are doing the coolest research in the department because [flattering comments]. I’m looking to get some engineering work experience. Is there any possibility of getting a research job working for you? I’d be willing to work for free for a while to prove that I can make a real contribution.” If he doesn’t have anything for you, go down the list. This was more or less my process for getting my first MechE job. I can all but guarantee that someone will hire you.

If no one in the department responds to this, then do the same thing with engineering extracurricular activities (Mars Rover, super-mileage car, whatever)—work down the list until you find somewhere you can get some experience solving engineering problems.

These experiences are not a substitute for getting an internship (don’t just do research and assume that is as good as an internship—it is not) but they can help you get your first one. In terms of resume value, paid engineering internship > paid engineering work on-campus > unpaid engineering work > paid non-engineering work, so act accordingly. Also note that teaching assistant jobs are not really engineering work; I don’t think I have ever heard a good engineering problem-solving interview story from a teaching assistant job.

Work hard at writing your resume, and get help

So now you’ve got a good GPA and some engineering work experience, and you need to communicate that to the world using your resume.

I’m going to gloss over this quickly, because resume principles are simple. Emphasize your achievements (I made the company $100K by going above and beyond!), not your responsibilities (responsible for pushing papers around!). Use action verbs. Quantify things. Maximize space spent on your unique achievements, particularly on-the-job achievements. Minimize space spent on the obvious—yes, every student thinks they can use Excel, engineers take engineering classes, etc. Put your GPA and your intended graduation date front and center; the recruiter is looking for this info and will just be annoyed if it’s not prominent, even if it isn’t great. Use standard resume formatting. Stick to one page; engineering students with multiple publications, multiple internships, and multiple scholarships can fit it all on one page, and you can too. Read the top Google results for something like “engineering resume advice” and consider their advice. When you have done all of this, get a few knowledgeable people to mark up your draft. Student organizations sometimes do workshops for this sort of thing, your department may be able to help, parents or family friends may have good input, etc.

Now, with your resume in hand and great engineering stories to tell, you’re ready for the career fair?

NO! There is much more work to do before the career fair!

Do sophisticated research on companies of interest

There is a stupid idea out there in the world that “the career fair exists so that students can learn about companies.” But if you want a job, the career fair is not a place to learn—it is a place to perform. So you need to prepare.

Prepare for the career fair like it will determine the course of your entire career, which is very possible. Read the list of employers, so that you know in advance who is hiring. Pick your target companies. If employers are taking 4 resumes per interview, and doing 4 interviews per hiring slot, you’d better be having quality discussions with at least 16 target employers to get your 1 job. (Or better yet, more, because 1 expected-value-probability job is not the same as 1 actual signed job offer.) Research each target company.

(Please, please, never, ever utter the career fair suicide-phrase “so what does your company do?”)

Base-level research consists of Googling the company, finding out what industry they are in, and scanning through some of their stock HR-speak about “exciting opportunities to develop yourself, while working with a team of the best and greatest people, in the context of the best and greatest HR organization.” This is good. If you have not done this, you will likely disqualify yourself from getting an interview. But every other credible candidate has already done at least this level of research.

If that’s basic research, what does sophisticated research look like? For any public company, you should look at the company’s annual report and investor slides. Let me repeat, look at the company’s annual report and investor slides. Every publicly traded company will produce an annual report and investor slides that contain information like:

  • What is this company’s leadership most proud of? If you think it’s cool, tell the recruiters! They probably agree, and will be flattered.
  • What are the company’s main strategic projects and priorities? More flattery-fodder! Also, these supply ready-made questions for the recruiters. “Has your department had any involvement with [strategic project]?” can start an interesting conversation.
  • How does this company view its relationships with competitors? Is it trying to be a technology leader? A low-cost producer? The biggest? The best in a narrow niche? This can help you talk intelligently about the company itself, rather than just the company’s industry.
  • What parts of the company are the most profitable? How do the parts of the company relate to each other? Having a concept of this can help you understand how your recruiters fit into the business, and what you might do if hired, again contributing to intelligent conversations.

To really take it up a notch, could also read the most recent quarterly earnings press release, or even listen to their earnings call (or better yet, read a transcript), where company high-ups reviewed the quarterly results with the investment community.

You can also learn about the company by talking to employees. Prepare for this appropriately, by reading all of the above stuff. Then ask recruiters, alumni from your school, friends of friends, etc. who work for the company if you can quiz them about it for twenty minutes sometime. Many people are pretty open to chatting about their jobs, even with random strangers. This can be a more frank and lower-stakes discussion than the moment when a recruiter says “any questions for us?” (That actually means “prove to me that you are actually interested in my company by asking me insightful, well-researched questions” and is not the best time to ask any question that could imply less than starry-eyed enthusiasm about the company.) One caveat: recruiters are busy during career fair week, and you should already be interacting with them at the career fair and info session. So plan ahead and do networking calls early, or find people other than recruiters to talk to.

Even for your 16th-favorite company, you should still spend a half hour looking at their investor slides. You need them to feel like you are seriously interested, and you will be if you have no other job opportunities! Doing half an hour of research on 16 companies is only one day worth of work. Not that hard.

Do sophisticated research!

Make the most of career fair week

So now you’ve got your resume tuned up and printed, you’ve done your research, and it’s time for career fair week.

The career fair is almost 100% about getting an interview. Hiring decisions are made based on interview performance, and interview decisions are based largely on resume and career fair performance.

Make sure you attend the info sessions for your target companies, which are often before the career fair! Participate during the info session. Stay after the info session, introduce yourself personally to the recruiter, give them a resume, and ask them one or two intelligent questions. Listen to their answers to other students’ questions.

At the career fair, work down your previously selected list of target companies. I’ve only been involved in one virtual career fair (which didn’t seem to be very effective), and I’m hopeful that we will have in-person career fairs within a year or two. So I’ll give advice on for in-person events, although similar principles apply to virtual career fairs too.

Walk up to the recruiter with a smile on your face and a resume in your hand. Introduce yourself and hand them your resume before they ask (they are expecting it—no need to even say “here is my resume”). Express interest in their company for a specific reason, ideally “I am interested in your company because of [unique capabilities and positioning within industry]” rather than “I like [industry, including you and all of your competitors].” Show you have done your homework and actually care. Draw their attention to a few key points of your resume—maybe give a twenty-second summary of your best work experience. Answer any questions they have; some recruiters treat the career fair like a mini-interview, and/or will want more detail on certain experiences on your resume. Ask one or two intelligent questions about the company. Express your desire to have an interview. Say thank you for their time. Don’t rush, but do avoid monopolizing the recruiter’s time if there’s a line to talk to them; a two or three minute conversation is often about right. Do talk to other recruiters from the same company.

Consider practicing your personal introduction with your roommate or the mirror or whoever until you can do it smoothly. If ye are prepared, ye shall not fear. Watch other students, and imitate the better ones. Ideally you should attend the career fair as a freshman, when stakes are low, just to see how these interactions go.

Hopefully all of this effort impresses recruiters and gets you some interviews.

Study for interviews like they are tests

A job interview is like a final exam on your professional life up to that point. Study for it like an exam! (Maybe more than for an exam…I know how some students approach exams.) Practice for it!

Most engineering employers will use behavioral interviews. The theory of a behavior interview is that past performance is a good predictor of future performance, so if you tell an employer how you have acted in the past, they can predict how you will act in the future. Or if you have never acted in the past, they will wonder if you will ever act in the future—this is why you need relevant job experience, as discussed above.

There are only so many ways to ask someone “were you a good engineer at the last place you did engineering,” so most behavioral interviews end up being pretty similar. To prepare, Google “behavioral interview” questions/techniques/etc. and read the top 5 or 10 articles to get a sense of what the questions are likely to be, and the standard advice for giving a clear answer (e.g. STAR, talk about Situation, Task, your Action, and the Result). Make a list of common questions from those articles, write out your answers to them, and literally practice telling your story/answer out loud until you can do it smoothly. Seriously, practice!

Since many of the questions are variations on themes, if you have a dozen really good behavioral interview answers, chances are at least one of them will fit almost any question. Your answers should probably be 3-5 minutes long, with a focus on concrete details (no hypotheticals) and things you personally did (no “we”). Some interviewees fell the need to tell me what they learned from an experience; I would generally prefer more detail about what they did.

Also think about whether the company in question has any specific values or needs that might come up in questions other than those on the standard list, and what stories you might tell related to those. For example, safety is very important to Valero (this is obvious from our website, info session, etc.), and thus our engineering job interviews pretty much always include a question about safety. Wouldn’t be a bad idea to ask older students what kinds of interview questions they have received…the types of questions you get will probably be similar, and even the exact wording of a given company’s questions might go unchanged for years.

Classwork-related answers can be OK, but are often weak. I think one day, more than half of the people I interviewed used the same group project as one of their stories; hard to stand out from the field with class-related stories when everyone else has taken the same classes. This is a huge part of why you should put so much effort into gaining engineering work experience: so you can have good, unique interview stories about actual engineering work. When looking for a job, you should ask: “is this a job where I will be able to get good engineering interview stories?” (Or will I basically be doing rote tasks?) And on the job, remember that being an outstanding employee will often result in a good story, and similarly, thinking “what would I want to tell an interviewer that I did” can help you act like a great employee. So get and use non-classroom engineering stories.

Anticipate a question about your geographic preferences. Remember that companies do not have a lot of flexibility; they are usually trying to fill specific positions in specific locations, not create a job just for you. Consider the long run; employers may be looking to get interns who they can then hire long-term in the same location, so going somewhere you hate “just for an internship” may be silly for both you and the employer. At the same time, you can and should be open-minded. Good job experience in a dubious location beats homeless unemployment in your first-choice location. Be and act as geographically flexible as you honestly can be. Less is more: “I am very flexible” is music to a recruiter’s ears, and whatever you say is probably going to be condensed into a yes/no or single sentence summary anyway, so don’t load them up with a lot of ambiguous information. “I am very flexible, and I’d be especially enthusiastic about [location] because I have [family/connection/interest/etc] there” could also be appropriate, or “I am very open to all of your locations except California.”

An interview will generally end with a chance for you to ask the recruiter questions. Prepare two or three smart questions in advance, based on your sophisticated research (see above). Do yourself a favor and do not ask “what are you looking for in an intern?” One of the number one qualities most employers want in an intern is literally the ability to understand requirements without having them explicitly stated in great detail.

Do ask the recruiters for their contact information. Send them a short email thank you note, ideally within a few hours of your interview. They may be making hiring decisions before they leave campus, so to the extent you want to impress them, a paper thank you note will likely be too late.

I got an internship—now what?

Purchase and read the book “The Unwritten Laws of Engineering.” You maybe also be able to find some or all of the content online. I was given a copy as an intern, I’ve since have read it multiple times, and I made it standard-issue for my new engineering hires. I do not expect to write a blog post on how to be a successful engineer, because this book, published in 1944, is still the authoritative word on the subject; the principles are timeless.

Wow, this is great advice, thank you!

You’re welcome! Feel free to share this post; some of this is zero-sum (if everyone formats their resume better…) but a lot of it is not. If every engineer ends up better-prepared, notably by seeking engineering experience early, that should actually improve the world’s standard of living. If engineers did better research into companies, they could end up better-matched with their employers, again resulting in welfare gains.

Congrats on making it to the end. Rant over.

Refinery Inspection, Putting Your Eyes On It, and Your Eternal Soul

Let’s tell a completely fictional story that bears no resemblance to actual refinery events.

Once upon a time, there was a refinery desalter. It was a fairly good-size pressure vessel, perhaps 15’ in diameter and 100’ long, in which water and salt settled out from crude oil before it went through its first distillation. This isn’t a spot in the refinery you would normally expect to see a lot of corrosion; it isn’t particularly hot (so immune to various mechanisms that require a lot of temperature/activation energy), everything is liquid (no possibility of preferentially condensing nasty stuff out of a vapor), the contents are relatively benign, and so forth. And on top of its required minimum thickness—something more than an inch of steel—the vessel was built with a healthy corrosion allowance, or so its documentation claimed.

One day, after this vessel had been in service for some years, a contract inspector was tasked with using an ultrasonic thickness probe to measure its thickness. He turned in a written report stating the thickness of the vessel was X. Unfortunately, on comparison with the vessel documentation, X was not just less than the expected thickness based on any plausible fabrication error or subsequent corrosion rate. It was also less than the required safe thickness for the vessel.

Fortunately, me the company engineer and especially Dustin Loden the company inspector had not been born the preceding day. So they told the inspector his result didn’t make sense and sent him back out to look again. Lo and behold, the vessel (reportedly) got thicker! But still not thick enough to be quite reasonable. The inspector was sent back out again, with his supervisor, to check. On the third try, the thickness turned out to be consistent with the vessel having minimal corrosion, as expected. Subsequent internal and ultrasonic inspections also found minimal corrosion.

One thing that isn’t part of the story is what the contract inspector did to get the numbers he reported. Perhaps the first time he wrote in what he thought was a plausible thickness while sitting in his truck; it’s comfier there! Then the second time he actually tried to measure it but failed to adjust for the temperature of the vessel, and the third time his supervisor did it for him. Perhaps he accidentally measuring the thickness of the wrong vessel. Perhaps he fat-fingered the wrong number when entering his original report. Perhaps the first time he did his level best but had an equipment bug, the second time he recalibrated his equipment, and the third time he better interpreted his machine’s readout.

It is surprisingly hard to establish the exact truth, even on relatively simple questions. The story above is just one example; reporting maintenance progress is fraught with related issues. The foreman tells his company superintendent that the welding is 90% done, an hour or so before shift change. Then the superintendent tells the refinery coordinator it ought to be done at shift change. Then the coordinator tells the project manager at the shift change meeting that “it’s done about now,” and the manager tells the world “it’s done.” Meanwhile a problem arose half an hour before shift change and now there’s still a shift of work to go. Then the night shift spends all night trying to get to where the day shift said they already were, and in the morning, it’s clear that the schedule has slipped and we’re six figures behind where we thought we were.

Then there are other inspection problems. One great thing about radiography (think X-rays, although in practice gamma rays are used) is that other people can look at the film an inspector generates and see for themselves whether a weld looks good, although expert interpretation is still required. Some of the fancy ultrasonic methods (shear wave, phased array, etc.) suffer by comparison because the operator can’t always readily show others the basis for his assertions. Some of the ultrasonic outputs can be hard to interpret, and the interpretation may rely on how the image changes as the inspector moves the probe, which is hard to document. So any time a shear wave tech sees a high-consequence problem (you need to spend $100K repairing this), the first response is likely to get a second opinion, ideally from “the guy we trust.” (There may only be one.)

Even when an inspector looks at a vessel, says “it’s bad,” and brings back a picture that looks bad, the case is not closed. Lighting and other aspects of the photography make a huge difference to the final image, and “bad” is relative.

In all these cases, it takes work to get your arms around the truth. (As good economists know, information is costly.) The dynamic is like the fog of war. Whether you’re working against men, their creations, or nature, a novel problem with time pressure is going to involve a struggle for truth and understanding. Novelty means you can’t run a pre-planned response and don’t have complete information, and the time pressure demands immediate action despite this. In such situations, lots of information is conveyed man to man, and its quality is uneven.

It’s the sort of thing that can make a man skeptical. And in fact it does generate a premium for firsthand knowledge. People qualify reports about field conditions or field progress by stating whether or not they have “put their eyes on it.” Good engineers and company inspectors are constantly going out to see progress or problems for themselves. “Trust, but verify” is the order of the day.

So now, shifting gears, consider: where else are these dynamics at work?

Well, this is also how your life works. Life is a novel problem (YOLO) with high stakes (your soul, if any) and time pressure (death has likely come a few minutes closer since you started reading this, although lifestyle changes from taking my later claims seriously will hopefully offset that).

You will probably be all right whether you decide on cookie dough ice cream or mint chocolate chip; life’s stakes are not always high. But they often are. Marriage is high-stakes; having a happy 60th wedding anniversary (congrats to my grandparents) is quite a bit different from divorce as a lifestyle (shed a tear for Nicolas Cage). Likewise for your basic attitudes toward life; despair and spiritual confidence lead very different places. If you have even the tiniest suspicion that God might exist, or even expect something out of you, that’s another good high stakes question to figure out.

In all of these cases it doesn’t take a genius to identify the goal: fix the refinery and profit, have a happy existence. But you need a lot more information to get there from here. The default pathway is to stay where you’re sitting, complacently accept some kind of lowest-common-denominator view, and see how things turn out. Your couch has all the spiritual answers you need to somehow muddle through to old age. This leaves spiritual ‘“profit” on the table, but some people seem to be fine with that.

If you want to really influence things for the better, though, you will have to do some work and find out for yourself how things really are. Then you’re left deciding who and what to trust. It may be a lot of work to go “put your eyes on” the truth, figuring it out for yourself, to your own satisfaction. But a good faith effort to dig into an issue goes a long way. Not only can you discover the truth; you can learn who already had it—who you can trust in the future.

So enough with hypotheticals, and on to conclusions where I have personally made the effort and I’m satisfied that I have the truth—I have “put my [spiritual] eyes on it,” as it were. There is a God and he expects things from us. He has said “Ask, and it shall be given you; seek, and ye shall find; knock, and it shall be opened unto you” (Matt. 7:7) and this is true; truth is there for the getting and will be revealed by the Holy Spirit to those who faithfully, sincerely look for it. Jesus Christ is the savior of the world, and the Old Testament, the New Testament, and the Book of Mormon are his word. I’ve studied all of these many times, and I have felt peaceful assurance from the Holy Ghost that they are true. Based on that, I know that Christ loves us, understands us, and provides guidance for us, not least through his authorized church, The Church of Jesus Christ of Latter-Day Saints. And thus I have pretty clear guard-rails for my high-stakes life decisions, and I can get further knowledge through prayer and study as I need it.

I have spent two years preaching this (as a missionary) and 10% of my lifetime income (tithing) on supporting its spread. Consider, further, that I voluntarily sit in meetings (for free!) in support of teaching this. But unfortunately, spiritual knowledge is not like X-ray film—no one else can look directly at my mind and see whether the Holy Ghost has confirmed the truth of the Book of Mormon, for instance. That is my report, but if you really want to know for sure, you would need to “put your eyes on it” by reading and praying about it yourself. Please do. This would be a good place to start. Or better yet, talk to me; I enjoy answering questions (yes, work folks, I’m not just that way about refinery stuff). I can also put you in touch with your friendly social-distanced local missionaries, who I think are now on every video chat system known to man.

All of which is to say: never trust anything anyone in a refinery says, except for me, this one time, about religion.

Saudi Arabia, 2040

Saudi Arabia will be interesting to watch over the next couple decades.

Oil prices have been crazy lately, between the whole Saudi-Russia dispute and COVID-19 demand destruction. Zooming out and looking past all that, I think the long view of oil market dynamics has interesting implications for the future of Saudi Arabia. Similar historical cases suggest a number of ways this could go.

In a competitive world with short oil project life-cycles, the cheapest sources of oil would be exploited first, and oil prices would tend to equilibrate out at a level where producing the most expensive barrel had a small economic return. Oil prices could move significantly as the most expensive barrel changed. But eventually, at any given level of OPEC/Saudi production, there will be an equilibrium price where supply and demand balance at that level of production.

It seems like $55 oil has been about right the last couple years. It keeps the major multinational oil companies doing frontier development, it keeps fracking going, it allows oil sands operations to turn a profit, and so forth. When more super-low-cost oil suddenly comes on line, whether because of geopolitics or geological surprises or whatever, prices and marginal fracking will decline to accommodate the cheap barrels. A drop in demand (COVID-19!) has a similar impact, shifting the world along the geological cost curve. But within a few years, natural decline of the world’s existing oilfields will tend to push the price upwards until typical projects, particularly fracking, can again offset natural decline.

It feels like there is a lot of supply elasticity around these prices, meaning that a relatively small increase in price will eventually result in a pretty large increase in output. If the market was in equilibrium and Saudi then decided to produce +2MMBPD vs. status quo forever, there would initially be a price bloodbath, resulting in reduced investment in projects across the globe, but after 5 years the marginal barrel might well end up being a fracked barrel at $53/BBL (vs. $55 if Saudi had done nothing). If the market was balanced and Saudi suddenly reduced 2MMBPD, there would tend to be a spike in prices, with correspondingly frantic activity in the Permian. But at the end of 5 years, marginal barrel would still be a fracked barrel, maybe at $57/BBL. Obviously these are cartoon numbers, but it is striking how many oil companies quote budget price sets or breakevens in the $50-$60/BBL range, and how much fracking output has been able to expand with prices in that ballpark.

With genius-level subtlety and planning, Saudi could potentially exploit the dynamics of the situation and yo-yo the market back and forth, in such a way that the price spikes filled their coffers whilst the dips keep fracking at bay, and mid-term average prices work out above a fracked barrel equilibrium. However, short fracking project cycles and American firms’ ability to hedge output make this harder. Long-run, Saudi is only likely to get $50-60/BBL, no matter what they do, and to the extent they act optimally they will probably tend to end up at or close to maximum production.

If the elasticity of supply was much lower, or supply slower to adjust, might be a different story. Around 2010 (pre-shale), the marginal barrel might have been an oil-sands or deepwater mega-project that took 7 years to come online. But that world is gone. (Or at least that is my impression, and my lazy Googling did not find credible recent estimates of long-run oil supply elasticity, so impressions are all I’ve got.)

So what does Saudi look like 20 years from now at $50-60/BBL?

Their budget needs $80/BBL to break even, so they will never again have a balanced budget without significantly higher taxes or lower spending. There doesn’t seem to be much to tax aside from the oil. They can go on a long time running up their debt to some high percentage of GDP, but sooner or later the money will stop, and they will have to cut back and start acting more like a normal country. To the extent the social contract is based on distributing oil money, that will be hard to do.

The situation reminds me of Argentina, where natural resources (under-exploited ranching land) generated great wealth for a period of time in the early 20th century. It was in the top 10 richest countries for a time. But ranch land eventually stopped being such a jackpot. Meanwhile cultural and institutional factors had remained sort of typical for the region, so the post-resource-wealth economy dropped down the global league tables until it also reached a near-typical level for the region.

So what does “typical for the region” look like for Saudi, neglecting the oil jackpot? Consider its neighbor Yemen, which had a 2018 GDP of $895 per capita. That would be a long way to fall. But of course Saudi is trying to diversify, and rightly so. Historically successful diversification has not been the typical experience in commodity-rich countries, or in Saudi specifically, but hopefully they break the mold. I would love to buy some cheap Saudi manufactured goods to go with my cheap Saudi oil.

However, assume for a moment that Saudi keeps struggling to diversify, and continues under fiscal (and thus political) stress.

Could Saudi go the way of Venezuela? In this case, shrinking oil wealth upsets a high-spending social contract, political repression results, the oil company gets bled dry under management of steadily deteriorating quality, and production declines through the years until a former powerhouse fades into irrelevance.

How about Libya? There, internal strife has resulted in on-again, off-again production as conflicting groups seek to control key infrastructure. This is harder to imagine; Saudi production (unlike Libya’s) might be “too big to fail,” and the U. S. tends to take an interest in it. But if production drops, Saudi could eventually become like Iran; a country that produces a lot, but remains dispensable.

On the other hand, maybe MbS can orchestrate an economic miracle and turn Saudi into Singapore-in-the-desert, a country that leverages a few initial advantages to achieve world-beating economic success. This looks like quite a challenge. It’s hard to imagine putting much non-oil-linked manufacturing in Saudi; wages are too high. Megaproject-based economic development is one of the knobs MbS really can turn, but there’s no straighforward path from megaprojects to an organic, vibrant economy. But MbS does seem to acknowledge the challenge he’s facing and has his “Vision 2030” plan to meet it, so let’s hope he can make it work.

Perhaps the best hope is that Saudi can just muddle through. Many of these same points could have been made in the 1980s, when Saudi was also running huge budget deficits, yet here we are, so maybe things just keep on going in a comparable vein. Maybe oil prices do get back to $80+ in the long run. Maybe MbS can gracefully trim spending. Maybe diversification goes better than we might expect from history. As at least some of these issues shake out the right way and things go on being OK, compound growth goes onwards, and the country prospers.

Cultural change is also a wild-card, although a fair degree of continuity is a good bet. Elsewhere on the spectrum of conceivable results, maybe a growing group of American-educated Saudi royals eventually brings the American Way to a populace that has decided, after decades of marinating in global internet culture, that they are just about OK with that. Fukuyama’s End of History finally arrives in earnest, and in 2040 we all sit around watching Marvel’s attempt to do for Arabia what “Black Panther” did for Africa. Maybe the trend of secularization slowly deepens and Wahhabism ends up sort of like the Church of England.

All of this is to say that I don’t know what will happen, but the potential variance in outcomes seems pretty large. (Much more so than for, say, Sweden—official 2040 slogan, “still a pretty good place!”) It’ll be interesting to watch.