Mongolia on Robert Ritz

Responsibilism - A better way to measure collectivism, especially for Mongolia

Fri, 25 Sep 2026 14:51:50 +0800

For years I heard about how some cultures are more “collectivistic” than others. A collectivistic culture is a culture that prioritizes the needs, goals, and well-being of the group over individual desires. Oftentimes I would hear about East Asian cultures being more collectivistic, with Western society being more individualistic.

Directionally this always seemed correct, but when you actually try to measure this, you can see that things start to fall apart. Let’s give a few examples:

“I put the group before myself.”
“I enjoy cooperating with other people.”
“Collectivists care more about other people.”
“Collectivists value harmony.”

These statements sound good on the surface, but are more about “warm fuzzy feelings” and aren’t actually very specific. Are we talking about my family, my coworkers, my parents, my country? Which “others” do collectivists care more about?

Historically surveys of collectivism relied on these types of questions to measure collectivism, and the result has been a confusing mix of answers. The concept of collectivism makes sense, but the way it has been measured never has.

A recent paper published in Nature Human Behavior that I had the privilege of helping with attempts to refine this understanding of societies. This new concept is called responsibilism. My contribution to the project was collecting data from Mongolians (n=194).

Responsibilism emphasizes responsibilities in close relationships. The Responsibilism Collaboration Team (snazzy name right?) collected survey data from 12,247 people in 100 cultures around the world. Instead of asking vague questions about “the group” or “others,” the survey used specific relationships and concrete situations.

Participants were asked to consider questions such as:

Should we keep our aging parents with us at home?
How much should we trust or help people we barely know?
How willing are we to form new relationships with strangers?

The resulting data helps us measure feelings towards duties and responsibilities in close relationships versus generalized warmth toward everyone. A simplified idea is who they feel responsible for. I may feel more responsibility towards my parents, and comparatively less responsibility to a stranger I meet on the street.

Across all cultures surveyed scores were given from 0-100.

Put simply a higher responsibilism score means stronger felt duties and responsibilities toward people in close relationships. In addition, a higher score is correlated with weaker felt duties and responsibilities at greater social distances.

This inverse relationship is shown nicely with these two examples:

Lying for a friend

Returning a lost wallet to a stranger

For societies with higher responsiblism, people will do more for those closer to them. Inversely, they will do less for those farther away. Unfortunately Mongolia isn’t listed in the charts above because these questions were asked as part of different studies. The answers were plotted against the responsiblism data, and the correlations are quite strong. This allows us to make some predictions about where Mongolia would end up here.

Question	Point prediction for Mongolia	95% prediction range
Return a lost wallet	~38%	13–63%
Lie in court for a friend	~35%	9–61%

These aren’t exact number obviously, but the trend is quite interesting. I built an explorer to view the data for Mongolia and compare it to other countries. Mongolia is an interesting country because it has elements of both Western “individualism” and Asian “collectivism”.

Here are a few interesting results I found, and where Mongolia differs from the United States.

Here a “good” friend can be defined as someone supporting you or as someone who notes that you maybe didn’t do something well.

Americans are much less trusting of random people near ATMs.

Awards are perhaps less common in the US, and therefore less susceptible to the awardflation that Mongolia suffers from.

If you want to see all the questions and how Mongolia compares, you can access the full explorer here: robertritz.com/responsibilism

However, I think the takeaway is that this research gives us a much more precise way to think about something that “collectivism” only described vaguely. The real question is not whether a society values the group over the individual. It is which people we feel responsible for, how strong those obligations are, and how quickly those obligations weaken as social distance increases.

Mongolia's Parliamentary Attendance Dataset

Wed, 23 Sep 2026 13:47:43 +0800

Fun new dataset on data.mn! Parliament attendance in Mongolia is a perennial topic, and the Parliament of Mongolia now has an API and dashboard to check on the attendance of MPs. I collected the raw data from 2024 to now and displayed it in a nice scatter plot.

You can view the new dataset here: data.mn/en/data/m…

There are many reasons for MPs to be late or absent, including intentionally to prevent a quorum as has happened recently. Nevertheless it’s an interesting dataset.

Advice to young graduates in Mongolia

Thu, 17 Sep 2026 16:00:44 +0800

This is the time of year graduates start coming over and asking advice, especially those who studied data science at AUM. I wanted to make a post to give these students my top points of advice in writing.

1. Show off your work

Showing a portfolio of work, especially printed out, is a great way to impress an HR person, and the hiring manager they hand it off to. Add charts, explanation, and aim for a semi technical explanation without getting too technical, especially with machine learning projects.

Often hiring managers are looking for experience with specific domains (housing, finance, even cars) so sprinkle in a mixture of different topics in your portfolio. When discussing these projects try to gauge the technical competency of the person. Sometimes they might know a LOT about a domain, but might not understand the data sciency stuff. Other times they might know both. Assume they know their business and give a non-technical explanation of the project. Who, what, why, where, when are all good things. And don’t forget “so what?”

2. Don’t work somewhere just for the experience, especially audit

This is super controversial, because I actually think you should work in your early career with a focus on experience. But that shouldn’t be the only goal. You should also find a workplace that gives you room for advancement, treats you with respect, and gives you the proper tools.

I say this especially about audit. Audit in Mongolia is a race to the bottom project, often with fixed price contracts. I consider it unethical to have a fixed price audit of any kind. With fixed prices, audit companies will be pressured to provide a completed audit within a specified budget. What if there is more work? You guessed it, unpaid overtime and tight deadlines with questionable quality.

3. Your starting salary matters

Many former students don’t put a lot of emphasis on starting salary, as they simply want to work somewhere to increase their experience. This is a great goal, but it also creates an opportunity for companies to take advantage of you. When you leave this first job, many companies will base your new salary on your old one. Raises also happen usually as percentages.

If you can get 15-20% more for your first job, you will probably end up with a higher salary in 5-7 years. Don’t short change yourself.

4. Ensure the company gives you access to AI tools

If your company does not have a corporate Codex, Claude Code, or even Muse account for you, I would think twice about working there. Today a single experienced analyst using AI tools can do the job of 4-5 employees from 5 years ago. These tools can securely handle customer and company data (remember the CIA is one of Anthropic’s biggest clients). If it’s good enough for them, it’s good enough for your company. Zero data retention means the AI company never saves your prompts or the results.

Any company in Mongolia today hiring analysts, data scientists, or coders without these AI tools deserves to lose all of the best and brightest Mongolian minds. They simply don’t deserve you.

5. Turn overwork into your advantage

If you boss keeps coming to you to get more work done, this is an opportunity, not a problem. If you help them out now, they owe you one. Take a bit of pain for a few weeks to help them out. If they don’t pay you back, never help them ever again with anything. Business works with mutual trust. If your coworkers work within this system, they will value you, and you will value them. For those that choose to take advantage, remember them and never help them again.

That’s all I can think of for now. I’ll add to this later if I can think of more.

Can Mongolia Support a 100MW AI Data Center?

Wed, 16 Sep 2026 23:46:12 +0800

If foreign companies want to pay protection money, maybe

Back in July, I wrote about Mongolia’s ambition to create data centers. There, I wrote mostly about current PM Uchral’s comments about Mongolia’s various advantages with regard to AI:

Mongolia is a non-aligned nation
Data sovereignty
An abundance of land in Mongolia
Mongolia is land-“linked” (i.e., close to China)

Since I wrote that article, a new draft law has been submitted to Parliament titled simply “About Data” (Өгөгдлийн тухай). I originally wanted to write an article about the financial viability of a 100 MW AI data center in Mongolia. I went deep into costs, electricity prices, fiber capacity, and more. Then I read the draft law. I ended up with more questions than answers, and no clear way to get answers to those questions.

So rather than talk about a hypothetical data center, I wanted to comment on several provisions of this draft law. I’m relatively positive about the direction this law takes data policy in Mongolia, but I also feel it is important to give my perspective on where things aren’t as good.

Encouraging foreign investment…for a price

This draft law takes a very notable step toward creating protections for foreign data centers built on Mongolian soil.

13.2. Data and information of foreign entities placed in a foreign data center shall be inviolable, and except as provided for in international agreements, Mongolian authorities shall be prohibited from accessing, physically or electronically inspecting, seizing, or suspending the operations of the data center.

This is a pretty remarkable set of protections for foreign data centers. In addition, and perhaps with a nod to my previous post, Mongolia is attempting to address the issue that data will flow across the borders of Russia and/or China.

13.6. In order to ensure the inviolability of foreign data center transit traffic passing through the territory of neighboring countries, special negotiations will be held with neighboring countries and an international agreement will be concluded with Mongolia.

13.8. In order to ensure the safety and reliability of the data center transmission network and to prevent potential technical risks, Mongolia shall organize at least three geographically and technically independent and unrelated international Internet gateways.

If AI tokens are a new type of “electricity,” having those tokens pass through a foreign power might not make a lot of sense. Energy independence is something most countries put considerable effort into. With AI, being “land-linked” isn’t currently seen as an asset, but rather as a liability.

These protections come at a price. In fact, they have a literal price tag. Emphasis is my own:

13.12. A benefit-sharing agreement will be concluded with a foreign entity that has requested to operate data center services in Mongolia, and the agreement will include the following terms:

13.12.1. In the event that the special protection and inviolability of the data center specified in Articles 13.2 and 13.5 of this law is guaranteed, the payment for the inviolability guarantee shall be deposited in the state budget at the established rate;

13.12.2. In the case of providing data transmission and export services using low-orbit or other satellite networks, regardless of the infrastructure of the Mongolian Communications Authority, special service fees to ensure network spatial independence shall be paid to the state budget based on the amount of exported data;

13.12.3. If you create an independent energy source using renewable energy sources to power your own data center, the energy independence fee shall be calculated based on the amount of energy generated and deposited into the state budget;

13.12.4. If the data center is powered by natural resources and generates its own independent energy source, the energy independence fee shall be calculated based on the amount of energy generated and the greenhouse gas emission offset fee shall be deposited in the respective state budget until the carbon emissions are reduced to net zero;

13.12.5. Create conditions for Mongolian government organizations, research and educational institutions to use a certain percentage of the data center’s computing, backup and storage equipment free of charge;

That’s a lot of fees! The biggest problem isn’t the fees themselves. The real problem is that the amounts are not set in the law and will instead be determined by the government later. This creates the possibility that the methodology or rates could also change later.

Foreign investors require consistency and trust before they will be willing to invest. We saw what happened with the Oyu Tolgoi negotiations, which were repeatedly mired in disputes and accusations of corruption. A better approach would be a static fee arrangement that is publicly defined and cannot be changed for the duration of the project.

I am not aware of another country that imposes this particular set of fees specifically on foreign data centers. It is also notable that domestically owned data centers would not have the same fees.

Tokens as electricity, and national security

I did some back-of-the-napkin math on building a 100 MW AI data center in Mongolia. My estimate puts the investment at around $6.7 billion, including dedicated power generation. Any foreign entity investing this much money in Mongolia will want serious stability, very much unlike what has been seen in mining in the past. The capital expenditure breakdown is:

NVIDIA GB300 NVL72 systems: $3.17 billion
Networking and storage: $0.67 billion
Buildings, site, cooling, electrical, and backup power: $1.3 billion
Delivery, spares, and IT contingency: $0.50 billion
Dedicated wind, solar, and batteries: $1.06 billion

This would be a construction project with a very high initial capital cost, but a relatively modest land requirement of perhaps 30–60 hectares. It’s easy to see why Mongolia wants to attract this type of business. It is also easy to see why these businesses, in my opinion, will most likely stay away from anything approaching this scale in Mongolia.

If LLM tokens are such an important resource, akin to electricity or even labor, nation-states may prefer to keep this generation capability within their own borders. Perhaps Mongolia could be a destination for training open-source models or running inference on them. For this to happen, a clear value proposition will have to be made, with transparent taxes and fees that investors can rely on without fear of later changes or renegotiation.

Four Letter Domain

Tue, 08 Sep 2026 11:14:03 +0800

Last year on a whim I bought data.mn. I’ve been working with Mongolian data for nearly 10 years now, and it seemed like an obvious thing to get. The owner had been sitting on it for years and didn’t do anything productive with it. I wanted to build a single portal with all things data on it about Mongolia.

The government has a few open data portals already, but they are generally not very easy to use. Search is somewhat poor, there are no built in visuals, or they don’t provide a secure connection (see below):

These portals are:

opendata.gov.mn - general open data portal based on CKAN, currently SSL certificate expired
opendata.burtgel.gov.mn - legal entity registration and things like that

So I wanted to make something dead simple, has great search, and work with AI agents. What resulted is a blazingly fast website that just gives you the data with a chart, CSV file, or Excel file. Excel file exports show a shortened form often used as a statistical table. CSV exports are “long form” which is ideal for creating visualizations in PowerBI or other visualization tools.

For AI agents, if the server detects if an incoming user is a person or a bot (we welcome bots!). If it is a bot, it redirects to a text only page which gives a description of the data and a the data itself in text form. This allows you to ask a question from ChatGPT or Claude about Mongolian data and it can use real, up to date data to answer you. Without this AI chatbots usually default to reports from Asian Development Bank or other international organizations with usually out of date data.

Data.mn currently has 120 published datasets mostly from Mongolian government sources. I hope to add more private datasets to the platform, something that is relatively rare unfortunately. If you have any ideas let me know!

Serious fuel shortage in Mongolia

Thu, 06 Aug 2026 12:34:21 +0800

This shortage is showing the strength of third neighbors and the weakness of direct ones

For the past few weeks, there has been a shortage of fuel in Mongolia. This is mostly due to the Russian invasion of Ukraine and Ukraine’s serious hammering of Russian refineries and pipelines. An additional challenge coming up will be the annual winter maintenance at the refinery in Irkutsk, where most of Mongolia’s fuel comes from. That usually shuts the refinery down for two to three weeks and causes shortages each year during the back-to-school rush.

Here are Mongolia’s fuel imports from 2021 to 2026. I wrote about a shortage back in 2021, and that one seemed to show an emerging pattern. At that time, a brief reduction in imports in early summer caused a cascade that resulted in a shortage a few months later. This happens for two reasons:

Mongolia has a limited amount of fuel storage capacity (about 45 days at most) for AI-92, the most common fuel for passenger vehicles.
Supply from Russia is limited, even before accounting for the war in Ukraine.

Mongolia is building more storage tanks, but this will only put a Band-Aid on the real problem: supply. About 95% of all fuel in Mongolia is imported from Russia, mostly from one refinery. If that isn’t a supply risk, I don’t know what is.

Source: [Data.mn](https://data.mn/en/data/mrpam-petroleum-imports)

Supply shocks from the Russian side are a yearly occurrence, and the war in Ukraine has only exacerbated them. When I wrote about the shortage in 2021, I noted that the government should have had a pretty good idea, months ahead of time, that there would be a shortage. Yet there was seemingly little public activity by the government to address the issue.

Third-neighbor strength

2026 is no different. New fuel supplies from South Korea and China have been secured, but they will have to arrive by southern rail. This route uses a different rail gauge, adding cost and time to the import process. In addition, fuel from South Korea must first be transported by sea, offloaded at a port, and then taken by train to Mongolia.

South Korea has pledged to supply 50,000 tons of fuel products monthly. China pledged an emergency supply of 6,000 tons, presumably as a one-time shipment.

Notably, Chinese oil imports have gone down significantly during the Iranian conflict. This is something of a mystery, given that China’s fuel storage, much of which is outdoors and visible from the sky, has not drawn down. Some experts in this space believe China may have a “secret stash” of fuel created in preparation for a potential conflict over Taiwan.

Benzin.mn

I’m astounded by the number of direct-name domains in Mongolia: oboi.mn, meaning wallpaper; unegui.mn, meaning free, although it is used for classified ads; data.mn, my own domain, meaning data; and medee.mn, meaning news. There are a few others that I can’t remember right now.

Well, now there is benzin.mn. Benzin is the Mongolian word commonly used for gasoline, although technically it means benzene, which is not gasoline. The site is a crowdsourced map showing which stations have gasoline.

This morning, I checked and only four stations in Ulaanbaatar were reported to have fuel. This is on top of an odd-even license plate restriction and a maximum purchase of 50,000 MNT.

Oil, oil everywhere, and not a drop for my Prius

Ironically, Mongolia has a lot of oil. It exports basically everything it produces. Production appears to be declining, something I didn’t realize until I looked at the chart this morning.

Source: [Data.mn](https://data.mn/en/data/mrpam-petroleum-production)

Nevertheless, in 2016, India decided to invest $1 billion in an oil refinery in Mongolia. At the time, this was the estimate of the fuel it would be able to produce:

…the oil refinery will have a processing capacity of 1.5 million tons of oil. As estimated, the oil refinery will produce 560 thousand tons of gasoline, 670 thousand tons of diesel fuel, and 107 thousand tons of liquefied gas per year.

By comparison, Mongolia imported 828,000 tons of AI-92 in 2025. This would therefore be a significant step toward reducing Mongolia’s dependence on foreign fuel.

There are a few problems though. Oil production declined from 4.6 million barrels in 2021 to 3.7 million barrels in 2025. Doing the math, 1.5 million tons is roughly equivalent to 30,000 barrels per day. If the refinery operates year-round, which is fairly standard apart from short maintenance shutdowns. That amounts to 10.95 million barrels per year. This means Mongolia’s current oil production would utilize only about 34% of the refinery’s expected capacity and therefore produce less gasoline that it is capable of.

It’s possible that there is a lot of oil-production capacity that could quickly be brought online and that I’m not aware of. The existing numbers don’t look good, though. Part of me wonders whether India’s investment in Mongolia was more of a snub at China than an effort to actually help Mongolia.

Winter prospects

It’s hard to say whether this fuel shortage will continue through the winter. The promised emergency supply from China will do little to help. However, the 20,000 tons of AI-92 (out of the 50,000 tons total) promised by South Korea will put a meaningful dent in the average 72,000 tons of AI-92 imported each month.

In the short term, at least through September, I expect the odd-even rule to continue. The real risk in all of this is diesel fuel. If that runs low, food trucks can’t get to market, and trains can’t transport food or coal.

With COP17 coming up, desertification looks like a small problem.

US visa bond program is now permanent

Tue, 04 Aug 2026 13:54:21 +0800

The visa bond program for the United States started impacting Mongolia in April 2026. This program requires B1/2 visa applicants to post a bond of $10,000-$20,000 dollars and limits visa terms from three months to a year. This was a pilot program, and it was just made permanent.

It’s annoying for us, to say the least. At least for now, when my wife’s 10 year visa expires, she will need to go through this process. Since she was 18 or 19 she has gone to the US many, many times. In fact she lived there for 8 years, which is when I met her. We were even married there.

We go to the US to visit family, and to do business, every year. One of the largest jewelry trade shows in the US is in Las Vegas each year and she attends. In fact most of the purchases for her jewelry business in Mongolia come from US suppliers.

I understand the State Department’s reasoning for this decision. I’d argue that this program, as it is currently structured, will probably cost the State Department more money. For all those thousands of Mongolians who travel to the US regularly, and who don’t break the rules, they will now have to go to the US Embassy in Ulaanbaatar once a year (in the best case) and do a visa interview. It is going to overload the already stressed system they have.

It would make much more sense to be able to waive this program for those with a long and obviously strong visa history.

Back in 2015 when my wife and I were married, she had to get a visa to go to the US. We made no secret of the plan to get married in Texas, and we planned to go on our honeymoon and then come right back to Mongolia. The consular officer basically said that she wasn’t allowed to get married in the US, which isn’t correct for obvious reasons. But the reason they said this is because once you are married to a US citizen and in the US you can file for an adjustment of status and then stay. That wasn’t our intention of course, but for a consular officer it’s certainly a bit of a weird thing to hear from a visa applicant in a country where most people would love to immigrate to the US.

Anyways, I tell this story to point out that my wife has had ample opportunity over the past 20 years to immigrate legally. She didn’t. So this program makes little sense other than to create more hassle for us and the consular officers at the embassy.

Funny enough, all immigrant visa issuance from Mongolia are paused anyways. So even if we wanted to move our family to the US we couldn’t…

Tue, 14 Jul 2026 18:57:21 +0800

This is our fourth time coming to this part of Bulgan Aimag in Mongolia. This little Soum center called Khutag-Undur has a nice hill next to it. Mongolian summer is a brief respite from a harsh environment.

Mongolia Wants AI Data Centers. First It Needs Power.

Wed, 01 Jul 2026 10:26:26 +0800

Mongolian Prime Minister Uchral Nyam-Osor was recently on a panel titled “No Power, No AI” at the World Economic Forum in Dalian, China. He touted Mongolia as a place to invest in renewable energy. In the seemingly inevitable AI future, global power consumption will dramatically increase. I’m generally positive about this future, but I also believe governments (with the exception of China) aren’t really taking power demand seriously.

PM Uchral wants Mongolia to be a destination for digital infrastructure and renewable energy. In my view, this is partly an attempt to turn Mongolia’s biggest infrastructure weakness into an investment story: the country badly needs more power.

The PM’s pitch sounds good. Mongolia has land, sun, cold weather, and proximity to China. But AI data centers do not run on vibes. They need power, water, fiber, political certainty, and security. Mongolia has some of those things in theory. In practice, each one gets complicated pretty quickly.

These are comments made by PM Uchral, edited lightly for grammar and clarity:

If you want to work in Mongolia, you have a few advantages:

Mongolia is non-aligned. It’s a democratic country with a strong interest in becoming a trusted digital and energy partner.

Data sovereignty is one of Mongolia’s big advantages. Companies seek secure and strategically located data infrastructure.

Mongolia has significant land opportunity, with the government owning most of the land.

Mongolia is a landlocked country. Mongolia is a land “linked” country and is close to China. You can link to China.

After he said this, the moderator, Michael Wang from CGTN, asked a question:

Moderator: How fast do you think regulation and permitting might take? Uchral: It’s about the licensing?

Moderator: How fast is it? Uchral: It’s fast enough. We will try… My main goal is about red tape reduction.

Note: CATL Chairman Robin Zeng gave Uchral a round of applause for this.

I want to talk about each one of these supposed advantages and explain why they might not actually be advantages.

Non-alignment

Mongolia is non-aligned. It’s a democratic country with a strong interest in becoming a trusted digital and energy partner.

Mongolia is truly non-aligned. Its small size and location next to two serious powers, China and Russia, create an environment where Mongolia is everyone’s friend and no one’s enemy. This is great for economic ties. Mongolia’s famous Third Neighbor Policy aims to use states such as South Korea, Japan, and even the US and EU as a counterbalance to Mongolia’s neighbors.

Unfortunately, this desire to make everyone happy means Mongolia wants to keep its neighbors (and other third neighbors) happy. A few examples:

Mongolia’s main North/South rail corridor has been largely stagnant since 1991. It is still 50% owned by a Russian partner, and this partner refuses to allow expansion, connection to other rail lines, or real investment. An East/West rail corridor would significantly improve industry in remote Mongolian regions, but it is dead on arrival due to this situation.
In 2018, agents of the Turkish government abducted a director of schools associated with the Gülen movement in Ulaanbaatar. They succeeded in getting as far as the airport before being stopped by Mongolian authorities. The agents were allowed to leave, and while the school director was saved from abduction, the schools were later shut down at the request of the Turkish government.
The proposed 315 MW Egiin Gol hydropower plant has been effectively vetoed by Russia. Russia claims it would negatively affect water flowing to Lake Baikal.

These examples show that while Mongolia is indeed non-aligned, it is often at the whims of its neighbors, and sometimes its third neighbors. Mongolia certainly is interested in becoming an energy partner, probably because Mongolia’s electricity generation has lagged behind demand. For a data center investor, the issue is not whether Mongolia is friendly. It is whether Mongolia can protect a project when a neighbor, supplier, lender, or political ally decides it has a problem with it.

Source: Data.mn

Data sovereignty

Data sovereignty is one of Mongolia’s big advantages. Companies seek secure and strategically located data infrastructure.

Uchral’s premise here is that companies are seeking secure and strategically located data infrastructure. Perhaps he wants to make something like the digital equivalent of the “doomsday” seed vault in Svalbard. So what do companies want when it comes to data sovereignty? I would guess several things:

Reliable infrastructure to power data centers
Solid security to keep data secure
Affordable prices for renting servers or colocating servers
Easy ability to move data in and out

That’s about it. Here is how Mongolia stacks up:

Mongolia currently does not have enough electricity for its own consumption, and water in Ulaanbaatar, where nearly all skilled labor exists, is also a constraint. The OECD projects that Mongolia’s water demand could exceed current supply by 2040 or sooner, with scarcity likely to worsen in high-demand areas including Ulaanbaatar. The recent MCC-Mongolia Water Compact increased Ulaanbaatar’s available water supply by nearly 80%, which is good news, but it also shows the scale of the problem. Northern Mongolia has more water, but it lacks skilled workers. The large Oyu Tolgoi copper mine in southern Mongolia currently flies workers in and out of Ulaanbaatar on a shift schedule.
Mongolia’s only publicly listed Tier III data center, which guarantees reliability, security, and redundancy, is run by Khan Bank, Mongolia’s largest commercial bank. Mongolia’s government-owned National Data Center claims to have a Tier III rating but is not listed on the Uptime Institute’s website. Mongolia’s new national data center aims for a Tier IV rating, but will be located in the New Zuunmod area, which shares water resources with Ulaanbaatar.
Prices for publicly rentable servers in Mongolia appear at first glance to be in line with international pricing, until you look into the machine details and see that they are running servers based on chipsets that first appeared in 2010-2014.
According to the International Telecommunication Union (ITU), in 2023, Mongolia’s total bandwidth was 590 gigabits per second across all links (fiber and satellite). Utilization in 2023 was 66% in total, but during peak hours my understanding is that these links are nearly full. The projection was 750 Gbps in 2024, but even this wouldn’t be a lot for an AI data center. A gigawatt AI data center would need somewhere around 1 terabit per second (Tbps). One terabit per second is 1,000 gigabits per second.

This is not even mentioning the danger that Russia’s internet kill switch poses. It is unknown how a possible internet shutdown in Russia would affect the internet in Mongolia.

Significant land opportunity

Significant land opportunity, with the government owning most of the land.

Mongolia certainly has land, and PM Uchral is right that the state controls most of it. That makes land available in theory. Unfortunately, it does not make a gigawatt data center easy to build. A project of that scale needs power, water, fiber, roads, permits, local consent, and contracts that investors believe will still mean the same thing ten years later. Local consent from the herders who use the land is a real sticking point.

Mongolia can bring in outside expertise, and the Oyu Tolgoi mine is the obvious example. It also shows the risk. Oyu Tolgoi has gone through tax disputes, financing disputes, a 2022 “reset” with Rio Tinto, corruption investigations around earlier agreements, and continuing political pressure to renegotiate the terms. The mine is successful partly because the resource is enormous. A data center investor will have to ask whether power, water, and political patience are enormous enough too.

Oyu Tolgoi also shows how expensive “available land” can become once a project gets going. The mine spent millions of dollars on local development, including roads, schools, health facilities, power connections, water systems, and herder support programs. Under its 2015 Cooperation Agreement, Oyu Tolgoi contributes $5 million per year to the Gobi Oyu Development Support Fund for community projects in Umnugovi aimag. By 2019, that fund had invested $22 million in 179 projects, including schools, kindergartens, a health care center, a flood prevention dam, livestock disinfection, and other local programs.

Moderator: How fast do you think regulation and permitting might take? Uchral: It’s about the licensing?

Moderator: How fast is it? Uchral: It’s fast enough. We will try… My main goal is about red tape reduction.

The legal framework for this push into energy and data centers has not been finalized yet. Even when it is finished, I don’t believe there will be an answer to how long a project would take from the start of the permit process to breaking ground.

Mongolia is land “linked” to China

Mongolia is a landlocked country. Mongolia is a land “linked” country and is close to China. You can link to China.

This is perhaps Mongolia’s greatest advantage, and also its greatest risk. First, proximity to China means easy access to many of the technologies required for data centers, including networking, cooling systems, solar, and newer battery technologies like sodium-ion batteries from CATL. The main exception is state-of-the-art AI chips, particularly from Nvidia. Chinese chips are indeed catching up, but they aren’t there yet.

Chinese AI companies are currently competing by being more efficient via distillation, or by increasing the number of Chinese AI chips, which are less efficient, thereby increasing power use. In China, increased power use is not as much of a problem given that China’s power production is increasing 9.7% per year (compared to Mongolia’s average of 5% per year since 2011, with demand outstripping supply by 30% in 2024).

China also purchases nearly all of Mongolia’s current exports, which are mostly mining exports. Yet China would not be a realistic consumer of data center output in Mongolia. An optimist would say that Mongolia, with its massive solar resources (about 250 sunny days a year), is perfect for solar. Mongolia’s cold, previously a big negative for battery storage, isn’t as much of a problem with new sodium-ion battery technology, which retains 90% capacity down to temperatures of -40° C. Yet it is hard to understand why a Chinese company would build a data center in Mongolia, as opposed to at home, where it would benefit from low-interest government loans.

I’m generally positive about Mongolia’s ability to be energy independent in the future (it isn’t currently). One thing holding domestic energy production back is price. Mongolia’s current energy prices (256 MNT/kWh, or about $0.07 USD/kWh on average for homes) result in domestic energy production being unattractive for private producers. China also keeps energy prices low through a mixture of state investment and cross subsidies. CATL recently announced a battery storage facility planned for Mongolia in the range of 100-400 MWh. This represents 4-14% of Mongolia’s energy production shortfall in 2024. For comparison, for a 100 MW AI data center, Mongolia would need more than 2,000 MWh of battery storage for that data center alone.

This isn’t a problem for China, given massive state investment in Chinese energy production. Mongolia does not seem willing to put massive state investment into power generation. At current prices, according to my own back-of-the-napkin math, a solar PV + battery power plant has a payback period of 20+ years, beyond the life of the batteries themselves, making this proposition unprofitable.

If the end consumer of this data center capacity is not China, but someone else, there is always the geopolitical risk to consider. With all internet fiber lines leaving Mongolia running through either China or Russia, and with close physical access to Mongolia, running state-of-the-art AI data centers in Mongolia would be a risk to both intellectual property and, given potential export controls on frontier AI models, to the companies operating them.

Mongolia’s own National Data Center was hacked in 2018 by suspected state actors and was discovered by an outside research lab, which notified the Mongolian government.

Poor Mongolia, so far from the ocean, so close to China.

Mongolia may eventually become a serious energy exporter or data infrastructure hub. I would like to see that happen. But the current pitch skips over the hard part: Mongolia is still trying to solve the same power, water, and connectivity constraints that an AI data center would immediately make worse.

A ton of copper concentrate is a commodity. AI may be the new electricity, but companies and states care where this new power is generated. Mongolia has not yet proven its ability to be a good steward of investor rights or a free market.

Mongolian Meat Prices, Seven Years Later

Thu, 12 Mar 2026 08:00:00 +0800

In 2019 I predicted meat prices would drop. They did. Then things got complicated.

Back in May 2019 I wrote about Mongolia’s meat prices, dug into the data on exports, animal losses, and inflation, and even made some forecasts about my predictions on meat prices going forward. At the time beef was 10,904 MNT per kilogram in Ulaanbaatar and mutton was 9,777 MNT. I argued that exports weren’t the main price driver, that animal losses were, and that meat was actually getting more affordable relative to wages.

Seven years is a long time. Since then we’ve had a global pandemic, border closures with China, and the worst dzud since 2010. I wanted to revisit the original analysis with fresh data from the National Statistics Office and see what held up and what didn’t.

The roller coaster didn’t stop

As of February 2026, beef in Ulaanbaatar costs 24,581 MNT per kilogram. Mutton is 19,181 MNT. That’s an increase of 125% for beef and 96% for mutton since May 2019.

But the path there wasn’t a straight line. After I published the original post, prices did come down through the summer of 2019 (my forecast actually got that right). Then COVID hit in early 2020. Mongolia closed its borders, including with China, and exports dropped. Prices stayed relatively flat through 2020 and into 2021.

The real acceleration started in 2022. A combination of post-COVID inflation, rising animal losses, and a weakening tugrik pushed prices up fast. By 2023 both beef and mutton were roughly double their 2019 levels. And then the dzud hit.

Exports: still not the main driver

In the original post I argued that meat exports weren’t the primary driver of prices. The data since 2019 only makes that case stronger.

COVID cratered exports. In 2020 meat exports dropped to 18,769 tons from 33,193 in 2019. In 2021 and 2022 they stayed below 10,000 tons. Yet prices kept climbing.

Then exports bounced back in 2023 to 34,194 tons (nearly matching the 2018 peak of 34,887 tons) and fell again to 24,736 tons in 2024. The correlation between exports and prices remains weak. Prices went up when exports dropped. Prices went up when exports rose. The relationship just isn’t there.

The dzud came back

This is the big story. In 2024, Mongolia lost 9.4 million adult animals. That’s the worst year since 2010’s devastating dzud when 10.3 million died.

But 2024 didn’t come out of nowhere. Animal losses had been creeping up for years: 2.6 million in 2018, 3.0 million in 2021, and then 4.9 million in 2023 before the 9.4 million in 2024. Two consecutive years of heavy losses (14.3 million animals across 2023 and 2024 combined) is something Mongolia hasn’t experienced in recent memory.

The total national herd dropped from about 71 million animals in 2019 to 57.6 million by the end of 2024. That’s a loss of roughly 13 million head in five years, or about 19% of the herd.

Looking back at the original post, I wrote that the 2.6 million animals lost in 2018 was “a very likely explanation for rising prices.” The 2023/24 losses were more than five times that number across two years. The connection between animal losses and prices remains the strongest signal in this data.

Meat prices finally outpaced inflation

This is where things changed from the original analysis.

In 2019 I showed that meat prices had been remarkably stable relative to inflation since 2011. For years, actual prices tracked below what you’d expect if they simply followed CPI. I interpreted this as meat getting cheaper in real terms.

That story flipped around 2022. Both beef and mutton prices broke above the CPI trendline and haven’t come back. Beef at 24,581 MNT is well above the 14,000 or so MNT you’d expect from CPI alone. Mutton tells a similar story.

The combination of back-to-back dzuds, post-COVID inflation (13.8% in 2021, 13.2% in 2022), and a weakening currency appears to have broken the pattern. Meat prices are now outpacing inflation.

Are Mongolians still spending less on meat?

Despite the price increases, the answer is still mostly yes.

I used the same methodology from the original post: a family of three in Ulaanbaatar, each person eating 250 grams of meat per day (roughly in line with average US consumption), earning the average UB salary.

In 2011 that family would have spent about 22.5% of their income on beef (or 19.7% on mutton). By 2019 it was 19.1% for beef and 16.0% for mutton. In 2024: 16.9% for beef and 13.7% for mutton.

The trend held. Wages in Ulaanbaatar have grown faster than meat prices. The average monthly wage went from 470,300 MNT in 2011 to 2,390,700 MNT in 2024, an increase of over 400%. Beef prices increased about 300% over the same period.

There was a noticeable bump in 2022 when wages hadn’t caught up with the post-COVID price spike, but by 2024 the long-term downward trend had reasserted itself. For the average UB household, meat consumes a smaller share of income than it did a decade ago.

The Tsuivan Index, revisited

This was a reader favorite from the original post. I created a price index using December 2010 as the base (100) for several common goods and services tracked by the NSO.

The results are pretty wild. Women’s haircuts have increased 885% since December 2010 (index of 985). Men’s haircuts are up 727% (index of 827). Both have far outpaced beef (505%, index of 605) and mutton (417%, index of 517). Canteen food (the NSO’s current category closest to tsuivan) sits at 494.

In the original post I noted that haircuts were the most surprising price increase. Seven more years of data just made that gap wider. Haircuts continue to outpace everything else in this basket by a large margin. This makes some intuitive sense: haircuts are pure labor, and labor costs in Ulaanbaatar have been growing fast. You can’t import a cheaper haircut from China.

What I got right and what I got wrong

Right: My forecast predicted prices would drop after the spring 2019 peak. They did. Animal losses remain the strongest predictor of price movements. Exports still don’t correlate well with prices. Meat has continued to get more affordable as a share of income.

Wrong (or at least different now): The inflation comparison reversed. In 2019, meat prices tracked below CPI. By 2026 they’re well above it. The 2023/24 dzud was a supply shock large enough to break the pattern that had held for a decade.

The question I raised in the original post about government reserve meat remains as relevant as ever. With 14.3 million animals lost across 2023 and 2024, the pressure on herders and on consumer prices is real. Whether the government’s reserve meat program can meaningfully buffer against shocks of this scale is still not clear from the available data.

All data in this post comes from Mongolia’s National Statistics Office (1212.mn).

Do families with more children make less money?

Sun, 30 Jan 2022 08:00:00 +0800

Any parent will tell you that having children is expensive. Raising children takes large quantities of clothes, food, education, toys, and the most costly thing of all, time.

Those same parents will also probably tell you their children are the best things they have made in their lives. As my wife has started saying, once you have children, you make every decision in your life with them in mind.

Women in Mongolia tend to have more children than in Western countries. In 2019 there were 2.8 children born per woman in Mongolia, significantly lower than the peak in 1965 of 7.5 children born per woman.

Let’s put this trend in perspective and compare Mongolia’s fertility rate with other countries.

We can see that it is a widespread trend for birth rates to have gone down over time since 1960. Yet around 2005, the trend in Mongolia started going up again. Mongolia today has a similar birth rate to China’s in 1979.

Other former communist countries show a similar trend. According to World Bank data, Kazakhstan, Russia, and Kyrgyzstan showed a decreasing birth rate that hit the lowest point around 2000. Since then, the birth rate in these countries has gone up.

I’m not sure what caused this “baby bump,” but it is certainly notable. I’ve been told by many Mongolians that increasing the country’s population is a matter of national security and that a high birth rate is a good thing.

When I asked why I was nearly always told the same reason. Since powerful, high population countries surround Mongolia, the country needs more people. This reason sounds quite lovely and patriotic. Unfortunately, at current birth rates, it would take hundreds of years to match even Russia’s population of 144 million, not to mention China’s 1.4 billion people.

So perhaps Mongolians have more children because of a sense of national pride or because they are freer than in the communist past. Perhaps what motivates some women is the “honored mother” medal to mothers with four or more children.

All of this got me thinking about the economics of having more or fewer children. It seems clear that families with more children probably spend less money per child or buy less expensive products than a family with fewer children.

But what about income? Do families with more children make less money? I dug into the 2019 Household Socioeconomic Survey run by the National Statistics Office to answer this. I calculated the average annual household income by the number of children. Let’s take a look.

Households with two children earn the most money on average. Households with five children earn roughly 40% less than households with two children.

Households with 0 children make the least. This is perhaps because these households are either relatively young or old enough not to have children in the house.

While this is a very clear picture, the cause isn’t clear. Do couples choose to have fewer children so both mother and father can work? On the opposite end, are mothers (or fathers) choosing the stay home to care for the children, resulting in a lower household income?

In 2020 Mongolia increased the child allowance to 100,000MNT per child under 18. It would put an average household with five children at nearly the same income as two children. The Mongolian government either wants to encourage people to have more children or support those families that choose to.

You probably have a pretty strong opinion of how many children are “right” for your family. Family planning is a very personal decision. The data reflects averages about the population and isn’t destiny for any particular family.

Mongolia’s current baby bump seems like it might deflate in the coming years. What remains is a very high percentage of young people. 31% of Mongolians were under the age of 14 in 2020. Overcrowded schools and traffic seem to be the result. We don’t know yet whether those born in the baby bump (after 2003 or so, mainly Gen Z) will end up better off than their millennial parents.

2021 Wrap-Up: Are Mongolians richer today than in 2011?

Fri, 31 Dec 2021 08:00:00 +0800

As 2021 ends, I’m reflecting on all the changes I’ve seen since I moved to Mongolia in 2012. As I’ve thought beyond the KFC’s, the CU’s, the new malls, and other surface level “developments”, I’ve been thinking of larger questions. One of the bigger of these large questions is a simple one, are Mongolians making more money than they used to?

It seems that every time you see statistics about average salaries in Mongolia, it seems to always be increasing. This was reinforced recently when the Lemon Press newsletter (link in Mongolian) noted that 6% of Mongolian make more than 3 million Tugriks per month (about $1,050 USD at the current exchange rate).

Their data, sourced from the National Statistics Office (of course where else?), also showed that nearly 25% of Mongolians make more than 1.5 million Tugriks per month. Lemon Press only looked at 2020 to 2021, and given my contemplative mood I wanted to compare a larger date range.

So, let’s do just that.

Do nearly 10 years make a big difference?

Before you see the chart, think back on your own life in the past 10 years. Ask yourself, have things gotten better or worse economically for you and your family over this time? I’m hoping the answer is better, and probably for most of you, it is.

One way to see how things have changed is to group households by an income bracket, 300,000-600,000 MNT per month for example. We can compare the number of households in each of these brackets over time to see if households are making more money than they used to.

The latest data available is from 2020 and goes back to 2011. Let’s see how Mongolian households have changed their income over these 9 years.

Wow, what a difference 9 years makes. My first reaction when seeing this chart was a good kind of shock. I was impressed by how this income distribution has moved up so much. Rather than a downward sloping graph where most households are in the lower bracket, more than half of households now make more than 900,000 MNT per month.

This is notable progress and it shows that Mongolia’s economy is rewarding those with marketable skills and entrepreneurial ability. There is always a good amount of criticism towards petty capitalism or corrupt officials, but I continue to see healthy businesses providing valuable services to Mongolians.

With that good news given, I want to bring up a point of concern with this data.

What about inflation?

Ah yes, inflation. That ever-present beast that gnaws away at your hard-earned savings. Inflation is often discussed in Mongolia, but often in terms of consumer prices or savings rates. Yet my point of concern is connected to spending power.

Inflation causes prices to rise, which is of course not good if your income is stagnant. If your income is going up, then you are racing inflation to improve your living conditions. Which will rise faster, your salary or the prices in the stores and the price of apartments?

In 9 years, from 2011 to 2020, the average monthly wage in Mongolia tripled. This is an impressive statistic, but what about inflation?

Let’s see how average monthly wages fared when adjusted for inflation.

To make the chart above I took the consumer price index (CPI) inflation measure for each year and adjusted the average monthly wage for that year. The results are less impressive than the income bracket chart above.

Instead of seeing wages triple over 9 years, we see a 59% increase. Put another way, if you made 1.2 million MNT a month in 2020, that is equivalent to about 670,000 MNT per month in 2011.

This is still very positive, and there have been many positive economic signals in these years. A major signal of this was in 2015 when Mongolia changed classifications from a lower middle-income country to an upper-middle-income country.

There are still areas of concern. What’s missing from the above data are those that choose not to disclose their income. This could be a large group of people on the lower end who don’t have consistent work. It could also be those at the very high end, whose income will come from investments in areas that are not a traditional job.

These blind spots in the data will be the areas average people get the most upset about.

So with 2021 wrapped up let’s look forward to the next several years.

What’s ahead for 2022 and beyond?

I won’t be making predictions, because that is a fool’s game. But I would like to make some observations about areas I’ll be looking at in 2022 and beyond.

Mining continues to grow and contribute significantly to Mongolia’s economy. The negotiations over Oyu Tolgoi are almost guaranteed to end up favorably for Mongolia, and Rio Tinto will be happy to have the dispute over with so the real business of money making can continue with the underground expansion.

The past few years have shown a big increase in the Mongolian Stock Exchange with both valuation and volume. Yet recent IPO’s have been underwhelming. The recent CU IPO is one example. It seems clear this IPO was designed to inject money into the business so the original investor(s) could have a nice exit and recoup their initial investment. Never mind a lack of profitability and a sliding stock price.

If 2022 brings more IPO’s of this type then the Mongolian Stock Exchange will lose whatever attention it started to receive from international investors.

I’m desperately hoping that Mongolia’s economy will diversify. Venture capital has proven to be the key ingredient in the United States’ success in innovation and tech. Yet the venture capital field in Mongolia is almost non-existent. Investors want high short-term gains. 9 out of 10 venture bets fail in the US, a statistic almost guaranteed to make a Mongolian investor stay away.

Even so, I’ve been impressed with the nascent display of startups, both tech and non-tech. This shows that young people especially are hungry to see the economy improve, even if the older generation isn’t always willing to help invest.

My wish for 2022 is for companies to realize that starting in Mongolia then expanding internationally is a difficult and perhaps losing formula. The Mongolian market and international market have few similarities, and the average European consumer wants very different things from the average Mongolian consumer. I would love to see companies start with the sole intention of selling their products or services internationally.

This would allow higher salaries that can compete with the mining industry and perhaps attract highly skilled workers who would otherwise leave the country to work abroad.

Welcome to 2022

I hope you’ve enjoyed the articles on Mongolian Data Stories over the past few months. I believe strongly that these explorations in Mongolian data can improve our understanding of the country we live in. After all, with the proper understanding, we can take effective action to improve our lives.

I wish you all a happy and prosperous 2022! Thank you for reading!

Which grocery store in Ulaanbaatar is cheapest?

Wed, 01 Dec 2021 08:00:00 +0800

Since moving to Ulaanbaatar in 2013, I’ve been to nearly every market in the city. I was struck by the opposite worlds of the zakh and the supermarket. One feels bustling but is arguably less convenient (no carts, very little parking). The other is more structured and has more packaged products but is often lacking in the selection of fresh foods. In some ways it’s indicative of the dichotomy that exists in Ulaanbaatar in general between the new and the old, the urban and periurban.

Zakh (зах) translates to market and is segmented in different areas such as vegetables and fruit, meat, and dry goods. In each area, vendors rent or own stalls where they sell their products. The products in each section are generally the same from stall to stall, with little variation. Cash used to be the only way to make a purchase, but more recently, card machines have become quite common (even preferred!).

A supermarket in Ulaanbaatar is pretty comparable to western supermarkets. Aisles of products with often many choices for each product. The main exception is the meat section. I’ve noticed that most supermarkets in Ulaanbaatar don’t stock much fresh meat, and if they do the selection is quite small (except Emart, which has a quite large meat section).

Even casual observers will note that prices in a zakh are lower than in supermarkets. Most of us only have a limited sample to work with, which means our observations are limited. Fortunately, the National Statistics Office of Mongolia tracks weekly prices for several products at all major supermarkets and zakhs in Ulaanbaatar. The products tracked are staple food products (meat, flour, bread, butter, eggs, etc.) that are most common in households. My understanding is this data feeds into consumer price index (CPI) inflation calculations based on the basket of goods identified by the government.

Don’t know what a “basket of goods” is? Imagine a shopping cart for the whole country. What set of products would nearly every family choose to purchase for their home? Probably basics like flour, eggs, milk, bread, meat, potatoes, carrots, and a few others. This set of products is then tracked to understand both CPI and can also be used to assess things like the poverty line, cost of living, and others.

You probably aren’t aware of this data source, as it is available only on the Mongolian 1212.mn page and also under a menu that might not grab your attention. If you are interested, check it out here.

I identified three questions I wanted to answer while looking at the data.

Which markets are most/least expensive over time?
Did COVID-19 have a noticeable impact on prices?
Do markets use psychological pricing? For example, carrots for 1,899 instead of 1,900 tugriks.

So, let’s get to it!

What markets and products do the data cover?

Here is a heatmap showing which products are available at each market around the city. Market names and product names were translated into English.

As you can see, not all products are always available at every zakh or supermarket. Goat meat and reserve meat are two notable items here. Goat meat isn’t always in demand in Mongolia, and reserve meat is only seasonably available depending on the Ministry of Agricultures program to provide low-cost meat.

I merged Mercury and Ikh Nayad as most of the shop keepers from the old Mercury market moved to Ikh Nayad. Also, Max Supermarket changed its name to M Mart, so I’ve reflected that in the chart as well. Saruul is a newer zakh, so there weren’t as many records for it.

From all the products in the chart above, I decided to select the eight common products (in my mind) to make the further analysis simpler and a bit easier to digest. They are:

Suu Brand Milk, 1L
Egg, Russian
Potato, Mongolian
Carrot, Mongolian
White Rice
Atar Brand Bread
Beef w/Bones
Mutton w/Bones

There will always be “one more thing” people may want to see in the analysis, but I hopefully covered the most common products. From here, let’s move on to looking at price differences between markets.

Which markets are most/least expensive?

If you live in Ulaanbaatar, I’m sure you already have an opinion about this. Something like,

“The meat is good at Bayanzurkh, but it’s cheaper at Bars.”

Or perhaps something more specific and even devious.

“Emart sells their vegetables for cheaper, but their meat is always the most expensive. But it’s great quality (Because obviously if it’s more expensive, it’s better quality!).”

I’ve never been to a country where people didn’t complain about prices in the store. It’s practically part of the human condition. Personal observations are always made with a limited sample, so let’s step back and see what the data says.

It was tough to figure out how to present this data. I generally prefer to show a time series to give the complete picture and let people decide. Even with only eight products, the time series graphs were confusing, and there wasn’t much of a clear point. I wanted to take a different approach.

A dumbbell chart allows you to see the difference in two times of several things all in one chart. You lose everything between those two times, but it gives a simplified picture that is much easier to digest than a line plot.

I created one chart for each of the eight products I selected. Each chart shows the difference in price from November 2019 to November 2021 for each market. This allows you to see how prices have changed and which markets are cheaper or more expensive between these two times.

Let’s take of look to see how eggs have changed in price over the past two years.

You can see that the price of eggs has increased. Supermarkets are also clearly more expensive per egg. The most expensive supermarket is 25% more expensive than several of the zakh’s.

I built a dumbbell chart for each of the eight products. Here is a gallery showing each of the charts.

After looking at these, you can come to some pretty easy conclusions. For most products, supermarkets are more expensive than at the zakh. Also, pretty much everything has gotten more expensive in the past two years. The exception is potatoes, specifically at Emart. For some reason, those got cheaper. Insert sinister conspiracy theory here.

I also think it’s worth it to look at the supermarket you usually shop at and ask yourself why you shop there? Are you sensitive to small changes in price, or do you choose your store because of convenience? If you are very price-sensitive, you now have some better information to make your decision. If not, well, then you at least know what the price picture is in other places.

One thing that surprised me is that Emart is not always the most expensive supermarket. It was my observation going into this article, and it didn’t seem to be true. Just keep in mind that our sample of products is limited to a few everyday staple products. Emart may keep prices low for these products to drive customers to the store and then charge higher prices for prepared or packaged food.

Let’s move on to the next question.

Has COVID-19 impacted prices?

There isn’t much to do here besides plot a time series of prices and take a look. Here it is. I’ve excluded meat prices as they are significantly higher than the other prices. The prices in the chart below are the average weekly prices of all markets in the dataset.

Considering that COVID-19 started in 2020 and the first serious lockdown happened in late 2020 in Mongolia, we don’t see much change in prices during that time. What we do see is a pretty significant increase in the past few months. In July 2021, carrots went above their maximum price from the previous two years. For the other products in the list, we see a clear upward trend since around September 2021.

There are many moving pieces to get products from the field to the market, so it isn’t easy to know what is causing the recent spikes. You would expect that you would see the market respond by adjusting prices after an increase in fuel prices or a market shock like a lockdown. This didn’t happen until very recently. I found this very odd.

AI-92 prices increased by 300 MNT per liter in July 2021, so there could have been a delayed impact. Combined with the Chinese border closure, these events could put more pressure on supply by increasing costs and restricting imports. Economists with more knowledge than me of Mongolia’s market dynamic could probably explain this better than me.

Lastly, I wanted to look at how markets price their products and perhaps see whether they use psychological pricing or not.

Do markets use psychological pricing?

Psychological pricing is when a business uses pricing strategies to either make the customer feel better about the product or make the customer believe it’s cheaper. For example, a product could be priced for 1,999 tugriks instead of 2,000 tugriks.

There are several psychological pricing strategies, and honestly, I didn’t want to test all of them for every product. One simple test is to count the frequency of prices that end in a certain number of zeros. For example, a price ending in three zeros would be 1,000, 4,000, etc. This tells us the price is an even thousand. We can repeat this test for one zero (‘0), two zeros (‘00), and three zeros (‘000).

Here are the results for three zeros (‘000).

From the chart above, zakhs tend to have prices in the thousands and supermarkets don’t. This intuitively makes a lot of sense, as markets may commonly accept cash, and items are purchased at individual stalls. At supermarkets, items are put into a cart and added up. So the price for an individual product is less noticed at a supermarket than a zakh.

Next, we can look at the frequency of prices ending in two zeros (‘00).

The story here is identical, not much to tell.

Finally, we can look at the frequency of prices ending in one zero (‘0).

The clear psychological pricing outlier of all supermarkets and zakhs in Ulaanbaatar is Nomin. Imagine a product priced at 1,479 vs 1,400. This doesn’t mean much to the consumer, but it could be a significant revenue boost across many products to the store. If 1,000 products are sold a day with an extra 79 tugriks added, that’s 2.37 million tugriks extra each month across many stores. That can equal a significant revenue difference.

Note that Saruul is lower here just because they didn’t have as many records in our data.

Did you change your mind?

After wrapping up this analysis, I can safely say that my mind wasn’t changed much. I mostly knew that supermarkets were more expensive. One thing I did learn was how similarly priced Emart is compared to other supermarkets on many products. Again, I don’t know if this pricing extends to products outside the sample we have, so this might not hold true in general.

What about you? Did you learn anything new with this data?

What happens if you pay people to get married?

Wed, 20 Oct 2021 08:00:00 +0800

We often think of marriage as a serious thing. When you marry someone, most people believe that it will be a lifelong commitment. From much research in many countries, we know that there are many benefits to being married, from living longer to lower insurance rates.

Therefore it makes sense that the government, and society, would want to encourage people to get married. So, what if you offered people the equivalent of $500 to get married? Would existing couples “on the fence” decide to take the plunge? Would people in need of quick cash make a sham wedding?

From 2006-2009, the Mongolian government did precisely this kind of cash giveaway for new marriages. Any couple registering a new marriage was eligible for this money, equivalent to roughly 4 months the average Mongolian salary from 2006. The results were nothing short of remarkable.

This situation is what social scientists (and economists) call a “natural experiment.” The government increased the incentive to get married, and we can easily see the result.

For the three years, from 2006-2009, the government offered cash to get married. In 2006 there were more than three times the marriages of 2005. For three consecutive years, more than 30,000 marriages were registered per year. No other year on record recorded so many marriages.

Economists are probably more interested in how the new incentive changed individuals’ decisions to get married. Social scientists might want to know how the institution of marriage changed, the impact on births, etc.

We can also break these marriages down by age group to see who decided to tie the knot.

Those 20-29 took the government money at the highest rate, with the 30-39 age group following them. If we look at growth rate, both age groups more than doubled their marriage rates. Notably, the 18-19 year-old group didn’t move much.

This government program has even spawned movie plot points in the Mongolian movie industry. The movie “Хуримын сюрприз” (Marriage Surprise) from 2015 is about a man wishing to marry the woman he loves, only to remember that he married his friend in college for the money. If you want to watch the trailer check it out here. It’s unclear if this is based on a true story (I’m joking obviously).

Take the Money and Run

As a socially minded data scientist, I’m more interested in how this large increase in marriages may impacted divorce. I immediately thought that a few years later divorces would spike while the people in it for the money decided to get out of their fake marriages.

Let’s take a look at the time series of divorces from 1988-2020.

This looks pretty terrible right? Beginning in 2005 divorces increased steadily. This chart alone probably reinforces the belief of many that a steady rise in divorces has been a negative symptom in society for the past decade. After all, a rise in divorces might mean more children growing up without two parents, more single men who are more likely to commit crimes, and a general erosion of a cornerstone of society.

As usual though, let’s dig deeper to figure out what is driving this trend. Luckily the National Statistics Office breaks down divorce by duration of marriage. Let’s take a look.

The main trend driving the increasing numbers of divorces seems to be marriages lasting longer than 10 years. Across all marriage duration categories divorces went down in 2020. Whether due to the pandemic or another issue isn’t clear.

Also notable are the smaller spikes we see for shorter duration marriages. I started getting suspicious seeing these spikes. They seemed to be staggered by duration. The 1-3 year duration divorces spike first, then the 4-6 year duration, and finally the 7-9 year duration. Coincidence? I think not.

To make things clear I highlighted the portions of each line where marriages would have happened from 2006-2009. This way we can see if divorces might have spiked because of the cash giveaway or some other trend.

It’s pretty astonishing how clear these spikes are. I find it nearly impossible to find another explanation for these divorce spikes. Clearly, this government program was responsible for a good portion, if not most, of the increase in divorces in the past 2 decades. Unfortunately, we don’t have data on the reason for divorce. Although I doubt any survey would include “we got married for the money” as the cause of divorce.

To extend this line of thinking further, we can relatively clearly see how divorces would not be nearly as high without these spikes. It’s reasonable to assume that annual divorces might be between 2,500 to 3,000 instead of more than 4,000 (a 25-35% decrease). Luckily divorces appear to be normalizing to the trend seen before the spikes.

Conclusion

As experiments go, the results are clear. If you pay people to get married, they will. Only years later they may get divorced, causing mayhem for statistics and making your grandparents believe that society isn’t what it used to be.

Each year there are between 10,000 to 20,000 marriages, and more recently, roughly 3,000 to 4,000 divorces. In the next few years, I expect that divorces will level out or even fall.

Whether this program was a mistake isn’t clear. The government intended to support marriage with this initiative. This seems to be a lesson to be careful what you wish for.

The rent is too darn high. Right?

Thu, 07 Oct 2021 08:00:00 +0800

It’s hard to think of something more at the core of society than the family dwelling. It’s where you hang your hat, put your family possessions, and spend time with the ones you are closest to. For many families in Ulaanbaatar, this means an apartment in a multi-level building.

As children grow up, get married, and have children of their own (but not always in that order), families often scrape together enough money to purchase an apartment for the budding young family to start their lives together. It sounds quite idylic, something akin to a “Mongolian Dream”, rivaling the American one.

It’s easy to argue that this Mongolian Dream is a false one, propped up by a highly profitable construction sector, but that’s not what I’m here to do. I’m here to answer a Tweet. That’s right, a Tweet. Perhaps you get motivated by lofty ideals or a sense of justice. I tend to pay attention to Twitter too much.

@RobertERitz do you want to take a crack at this? Median income is with NSO, average flat price index is with @TenkhlegZuuch and MIK group company
— Эрдэнийн ЛХАГВА (@Lkhagva) September 24, 2021

Following the Evergrande debt crisis, Bloomberg published an article showing the price to income ratio for various Chinese cities compared to well known cities around the world. I was asked to take on this chart and see how Ulaanbaatar stacks up with the rest of the world.

First let’s define how the price to income ratio was created. The data for this chart came from Numbeo, a living costs data aggregator. Numbeo uses a simple formula consisting of the ratio of average price per square meter and average salary. They assume an apartment size of 90 square meters and they take 1.5 times the average annual net salary as a measure of net disposable family income. This 1.5 number assumes an average of 2 earners in a family with 50% of the women participating in the workforce.

These assumptions might not apply well to Mongolia, but we can address these later. By having a consistent methodology we can compare all of these countries on housing affordability. Conveniently, Numbeo actually does calculate the price to income ratio for Ulaanbaatar. All put together, here is the chart from mid-2021.

Being towards the bottom of the chart, Ulaanbaatar does not look too bad next to the ultra expensive cities in China. But look at the two cities Ulaanbaatar is more expensive than in this list. New York and San Francisco!

Siri, recalculate that for me

Let’s take a step back. How sure are we that these numbers are accurate? Using local and reliable data sources, I wanted to calculate the price to income ratio to verify that the result from Numbeo is accurate.

Digging around for data, there are a few issues with the price to income ratio as calculated by Numbeo. First, it is not clear whether Numbeo is using average or median salary in its calculation. In Mongolia the median salary is substantially less than the average salary.

This is due to a small number of high earners skewing the average higher while the median (or middle) salary is lower. While average salary data are available for both Ulaanbaatar and the national level, data for median salary are only available at the national level from the National Statistics Office of Mongolia.

It is also important to note that generally when a median is much smaller than an average, the mode (the most common observation) is generally even smaller than the median. Given these facts we probably should not consider the statistically average Mongolian to actually be very average at all.

The second issue is related to the price per square meter of housing. In countries like the United States the average price of housing would include apartments for purchase, smaller multi-family dwellings, and single family units. In Ulaanbaatar, when the price of housing is discussed it almost always refers to the price of apartments, not single family units. Yet 50% or more of the population of Ulaanbaatar does not live in apartments. So any ratio we make with the available data would be limited in scope to apartments and not the full spectrum of housing available in Ulaanbaatar.

With these issues in mind we can calculate a price to income ratio using available data. I used the same formula as Numbeo to keep things consistent.

Average Ulaanbaatar salary for Q2 2021 via 1212.mn: 1,433,100 MNT
Average price per square meter of apartments for Q2 2021 in Ulaanbaatar via Doljoo from Tenkhleg Zuuch on Twitter: 2,647,076 MNT
10% Income Tax + 13.5% Social Insurance = Tax rate of 23.5%

Putting it all together we can take the average price per square meter multiplied by 90 square meters to get the average price of an apartment. Then we take the monthly net salary (average salary minus taxes) and multiply it by 12 months. Then multiply by 1.5 to get the estimated family income. All together:

The result? 12.07. This is a bit above the Numbeo ratio, but not by much. We should consider their number reasonably accurate given the issues we outlined above. Put simply, the “average” Mongolian family, which isn’t actually very average, will pay more for an apartment in relation to income than an average family in New York or San Francisco.

90 Square Meters? You rich bro?

If you keep up with things in Mongolia, you will know there is also an issue with using the assumed 90 square meters for an apartment. First, the government mortgage program allows a maximum size of 80 square meters. Second, a 90 square meter apartment is considered to be on the larger side for many in Ulaanbaatar.

To address these issues I decided to take another pass at the price to income ratio using more realistic numbers for Ulaanbaatar. Here are the data assumptions I used.

Median salary, national level, Q2 2021, via 1212.mn: 993,600 MNT
The same average price per square meter for apartments, Q2 2021: 2,647,076 MNT
10% Income Tax + 13.5% Social Insurance = Tax rate of 23.5%
Average apartment size: 62 square meters
- Tenkhleg Zuuch currently considers this the median apartment size in Ulaanbaatar.

The result? 11.99. Nearly identical to the ratio calculated above with average salary and 90 square meters. We can’t directly make a comparison now to New York or San Francisco like we did before. What we can do is draw a general conclusion about apartment affordability for a middle income family for a median sized apartment.

Note: You might think we could simply take the ratio from New York and multiply it by 1.3 (because 62 square meters is about 30% less than 90 square meters). Without knowing whether Numbeo used median or average salary in all circumstances it’s hard to know if we can make a generalized estimate like that.

But what comes next might be more useful.

Zoom in, enhance!

After doing the above calculations I felt that I still had not gotten the full context of the situation. I realized what we needed was a look at the trend of the price to income ratio. In other words, have apartments gotten cheaper or more expensive relative to wages over the past several years?

Luckily I had apartment price data and wage data from 2014-2021. Here are the data assumptions I used to gereate the chart:

Yearly net salary = Average salary (national level) x (1 - tax rate) x 12
Average apartment price = Average price per square meter in Ulaanbaatar x 62 square meters
Price to income ratio = Average apartment price / Yearly net salary

I was fairly shocked by the result. I always heard that families are being priced out of apartments, but this chart tells a different story. In the past 10 years apartments have become more affordable to both average and middle income families. Since around 2019 the trend has leveled off, with salaries and apartment prices rising at a similar rate.

This seems to signal that more recently supply and demand has reached some sort of balance, with the supply of apartments matching the demand in the market.

Does this mean that apartments are “easy” to afford for families? That’s a tougher question that leads us into mortgage territory and the unique properties of the lending market in Mongolia.

We can leave that for another day.

Mongolia is running on fumes

Tue, 21 Sep 2021 08:00:00 +0800

Just yesterday I learned that there has been a gasoline “shortage” for a few weeks in Mongolia. I put shortage in quotation marks because it seems the government and average citizens did not immediately agree on the reality of the situation.

On September 14, 2021, the Mineral Resources and Petroleum Authority of Mongolia made a statement via the state news agency Montsame unequivocaly stating that there is no shortage of gasoline. This is heavily contrasted by reports on social media of limitations on purchases or outages at some stations.

Хил гааль дээр хэдэн сараар гацсан бараа таваараа татаж чадахгүй, хэдэн колониуд нь бензиний нөөцөө барсан, ковидоор өвдсөн хүмүүст нь эмнэлэг олдохгүй, дарга эрх мэдэлтнүүд нь эзгүй.... бид дижитал үндэстэн pic.twitter.com/2xxAd3ha80
— Seku (@ganaasel) September 20, 2021

So what is the actual situation currently in Mongolia? As usual, I like to go to the actual data to see what the reality is.

Fuel Reserves

The Mineral Resources and Petroleum Authority of Mongolia (MRPAM) reports fuel reserves on a monthly basis, and has these reports going back to 2015. Unfortunately the reports are in PDF form, requiring a bit of copy pasting on my part to collect the last two years of data.

We can see that the reserves of AI-92 fuel as of June 2021 are nearly 35% less than June 2020. It is also clear that during the fall reserves are reduced, most likely due to increased vehicle usage as people return to work from their summer vacations.

As far as I can determine the government has not said anything definitive about when imports of AI-92 slowed or stopped. The government did release some information about further dwindling reserves today (September 21, 2021), stating the central region has 4 days of fuel remaining. Other than a mention that no AI-92 fuel will be imported in September, it is not clear what the current import situation is.

Note: I chose to visualize AI-92 fuel as it is the most common fuel used in passenger vehicles in Mongolia. It is also the type of fuel nearly everyone is complaining about shortages of.

Reporting Irregularity

Typically, the MRPAM posts the reserve statistics one month delayed, but it seems this year they have neglected to post any reserve statistics since July 22nd. By now we should be able to expect the reserve numbers for July and August to have been posted. As of the time of publishing, the government still has not released reserve or import statistics for July or August 2021.

In addition, an irregularity in the reporting popped up as I was going through these monthly reports. On all monthly reports, the MRPAM shows a yearly comparison on the fuel reserve chart, for example showing January 2021 vs January 2020. This allows the report reader to see how the reserves land year over year. This is a good comparison as consumer behavior follows a yearly pattern.

The latest report for June 2021 made a departure from this. Instead of showing a year over year comparison they compared June 2021 vs May 2021. This made the numbers look better on the chart, and conveniently hid the nearly 35% reduction in reserves year over year. The screenshot below from Page 25 of the June 2021 report shows this departure from the usual practice.

Fuel Reserve Projection

Given today’s announcement by the government, we know that the central region of Mongolia has only 4-5 days of fuel remaining. Nevertheless, it may be useful to understand how a fuel reserve projection could be made given imports, reserve numbers, and inferred usage.

MRPAM releases import numbers by fuel type alongside the reserve numbers.

A few minutes of copy pasting later I was able to create a nice table with reserves and imports for AI-92. I then took the previous months reserves, added it to the current months imports, and subtracted the current months reserves. This should give us a reasonable estimate of actual usage of AI-92 during that month.

We can also determine the available reserves at the end of the month given a very similar formula.

To extrapolate forward I looked at several different import scenarios where the imports are a percentage of the values from 2020. We will forecast forward using 75%, 85%, and 95% of the import levels from July 2020 to December 2020. We will assume the same level of fuel usage from July 2020 to December 2020.

Stepping forward each month we can do the process listed above and get a projection of AI-92 reserves at the end of each month for each of these scenarios. Keep in mind the projection below assumes 2020 levels of consumption and are based on a percentage of imports from the corresponding month in 2020.

At 85% of 2020 imports we see reserves emptying in September. At 95% reserves go negative in October and rise again positive in November. Only at 100% of 2020 import levels do the reserves stay above 0. The already lowered reserves from late summer seem to be exacerbating the current import issues.

It is very clear that without imports equal to or larger than 2020 Mongolia’s AI-92 fuel reserve will continue to be near zero for a few months. Of course these projections aren’t a guarantee of what will happen. They are merely scenarios based on past data that we can use to better understand the current situation. I hope you find them useful.

Hiding Risk

It seems fairly clear from the analysis above that the current shortage of fuel could have been seen as early as July/August this year. Reserves were so far off their usual level that it was a clear signal of risk. Of course this was a sensitive time politically as back to school preparations began. The government seemed to clearly be trying to exhibit a sense of stability and preparation ahead of the usual fall rush in the city and around the country.

It is important to remember that in December 2020 Mongolia had a large opportunity to significantly increase coal exports to China. China recently banned the import of coal from Australia, giving Mongolia an obvious advantage. Due to COVID-19 outbreaks and customs issues, Mongolia seems to have squandered that opportunity.

Mongolia's $150,000 license plate

Mon, 15 Feb 2021 08:00:00 +0800

Would you pay this much for the country’s “best” license plate?

TL:DR Mongolia auctions off a “lucky number” license plate and the bids reach 427.45 million MNT (Mongolian Tugriks) in less than three hours. That’s $150,035.10 USD at current exchange rates.

Numbers mean a lot to most Mongolians. Generally, 1, 3, 8, and 9 are considered “lucky” numbers. Mongolian phone numbers have 8 digits, and the first 4 of these numbers can tell you a lot about who is calling you. So-called “lucky” phone numbers beginning with 99 or 88 regularly sell for anywhere from $1,000 (9999-XXXX) up to $20,000 or more for the most elite (9911-XXXX). These phone numbers are sold on the “secondary market” from person to person. If a 9911-XXXX number calls you and you don’t answer, expect them to be upset…

Unfortunately, license plate numbers are assigned to a vehicle and can’t be transferred between individuals. In addition, up until recently, you couldn’t pay for a better license plate number. This caused a situation where the “best” license plates (think 0001-УБА) somehow ended up on Mercedes G-Wagon’s and Land Cruisers. I wonder how that happened?

The G-Wagon. Source: Wikimedia Commons

The general public always knew what was really happening. At one point frustration was so high that there was discussion of giving these lucky numbers to garbage trucks or other city vehicles. It seems their voice was heard when the new Motor Vehicle Registration Country Number Procedure was approved in late 2020. This procedure set up a system to allow “lucky” numbers to be auctioned on the Vehicle Registration Department’s website. This has been generally viewed as a positive development, as those willing to pay for this special license plate could pay legally for the privilege and contribute to government revenue and social programs.

I’m not sure anyone expected what happened next. On February 15, 2021, the number 0001-УАР had its day to be auctioned. The first bid was 1,000,000 MNT (approximately $350 USD). Let’s see what happened…

Cumulative bids for various license plates on February 15, 2021. Source: Author

The auction opened at 9:00AM. In less than 30 minutes bids for 0001-УАР reached the equivalent of $100,000. By 11:30AM the auction total reached 427.45 million MNT (Mongolian Tugriks). That’s $150,035.10 USD at current exchange rates.

Anyone bidding on a license plate number is required to put in a deposit for their bid of 50,000 MNT (about $17 USD). There were 957 bids for 0001-УАР. Those who didn’t win the auction will have their deposits returned. The winning bid is now expected to pay the total amount. If they don’t pay or are unable to, this number will go up for auction again.

It’s unclear how much of this auction was real interest or social media hype, as this particular license plate went somewhat viral on various Mongolian language Facebook groups. To put things in perspective, the average yearly salary in Ulaanbaatar, the capital, is about $5,300 USD.

What do you think? Is a $150,000 license plate worth it?

Is Australia's Loss Mongolia's Gain?

Fri, 18 Dec 2020 08:00:00 +0800

Note: This article originally appeared on Mongolia Weekly.

Typically, towards the end of each year China’s coal imports decline, only to restart with vigor early the next year. Argus media, a media organization focused on commodities, say that this slowdown, officially blamed on customs clearance issues, has happened for the past two years. Of course, this official explanation has shown a remarkable regularity at the end of the year when China has mostly satisfied its domestic need for coal. In both 2018 and 2019 around November, a line of coal trucks kilometers long would sit at Mongolia’s southern border with China in below freezing temperatures waiting to be processed.

As early as November, New Hope, and Australian coal miner, said they believed Australian coal imports were simply being delayed until January. Yet this year seems to be different. The recent trade spat between China and Australia has resulted in China eliminating clearance restrictions from several countries, except for Australia.

Since July, Mongolia’s coal export has been ramping up and trying to make up ground that it lost due to coal exports dipping to near zero in March. Mongolia’s Finance Ministry stated in 2019 that mining revenues made up 27% of the state budget. In addition, prior to just a few days ago Mongolia might have been looking at a hard slowdown of customs clearance for coal exports at the border in November. This race to the finish has not been kind, with total mineral products exports down $10.7 billion USD at the end of November, year over year, according to the National Statistical Office.

While the spat between Australia and China would seen to be an opportunity for Mongolia, it is important to keep in mind the type of coal that Mongolia predominantly exports to China. Since 2019, 85% of Mongolia’s coal export to China has been coking coal, which is used in blast furnaces as part of the steel production process. During the same period, only 42% of Australia’s export to China is coking coal, with a large portion of the rest being used for thermal coal to generate electricity and heat.

Nevertheless, for the past few years Mongolia has been running up against China’s de facto coal quotas. With Australia sidelined, Mongolia has a significant opportunity to ramp up coking coal exports.

In 2019 Mongolia’s share of coking coal exports hovered around 50% of China’s total import. In 2020 Australia filled the gap left by Mongolia and in March accounted for nearly 80% of China’s coking coal imports. Now it is Mongolia’s turn to fill the gap. If Australian coal imports remain stranded at sea, Mongolia could increase coking coal exports significantly in the near term.

Challenges remain for Mongolia. Currently coal is still trucked to China, and the under construction Tavan-Tolgoi Gashuunsukhait railroad is scheduled to be completed in 2022. This railroad will improve margins by reducing transportation costs and speed customs clearance, which will result in higher revenues and increase export ability. Until then, the main bottleneck is still customs clearance with China.

The recently opened “green channel”, which optimizes clearance of cross-border freight and related personnel, has been said to help. Yet due to the recent outbreak of COVID-19 in Mongolia, China insisted truck drivers entering the country be tested before being allowed to cross the border. Currently there is a big opportunity in front of Mongolia, and we will see early in 2021 if it can take advantage of it.

Mongolian Electoral District Analysis 2020

Wed, 17 Jun 2020 08:00:00 +0800

2020 Parliamentary Election

The current election saw a switch back to the block voting system used in 1992 and 2008. This system has both positives and negatives, and this system has much larger election districts than the previous single-member districts of the 2016 election. This allows us to more easily analyze the districts with the available data.

To put all of the data together and make it user friendly, I built an interactive data explorer that allows the reader to see key indicators of election districts. The indicators covered in the app include:

Population indicators: population, household size, herder percent, age group breakdown
Economic indicators (self-reported): salary, household loans, employment, education levels
Quality of life indicators: crime, house type, toilet & water type, heating type

The data was collected from 1212.mn and also the Household Socioeconomic Survey of 2016, which is the latest survey that contains precise geographic labeling. Both data sources are maintained by the National Statistics Office of Mongolia.

Several themes developed through my research into this data. These themes are perhaps not surprising for many readers familiar with Mongolia, but it is never a bad thing to back up observations with data. Below I give an overview of the themes for your convenience.

Ulaanbaatar vs Countryside

Perhaps the best-known difference in Mongolia is that of Ulaanbaatar and the countryside. The image of busy urban Ulaanbaatar juxtaposed with the bucolic countryside could hardly be more different. Indeed the data shows significant differences in many areas. Let’s compare the population pyramid for the election district with the highest population, Songinokhairkhan District in Ulaanbaatar, with the least populace, Bulgan Aimag.

There is an obvious population difference, but the size of the age groups also shows a marked difference between the two. Where you have a large spike in the 20–40 age group in Songinokhairkhan, Bulgans’ age groups have much less variability and are quite flat from 15–50. You will see a similar trend in the highest population election districts compared with the least populated ones.

This means the voting demographics in rural areas are more widely distributed than in urban areas. D.Byambjav and E.Gerelt-Od previously discussed the large and less engaged block of voters under 35. If youth engagement initiatives are successful, this block of voters will be less powerful in rural districts that have a similar age distribution to Bulgan.

Employment differs significantly as well between Ulaanbaatar and most rural districts. Rural election districts tend to show significantly higher rates of overall employment than the urban ones. In the chart above, the “employed percentage” refers to the proportion of the working-age population that is employed. Urban districts tend to have between 45–55% employment, with rural districts between 65–75%.

Salaries are also significantly higher in Ulaanbaatar. The salaries shown above are self-reported from 2016, but they show a very clear picture. Employment levels may be higher in rural areas, but salary levels are conversely lower. This is an obvious area where the economic policy of the party platforms could address these inequalities.

Outliers: Mining Areas and Bayan-Ulgii

The Ulaanbaatar vs countryside theme may be the biggest theme in the data, it doesn’t give the full picture. There are two sets of outliers that “buck the trend”.

The first is that the two primary mining areas of Mongolia, Erdenet (in Orkhon) and Oyu Toloig (in Umnugovi), show significantly higher salaries, lower loan rates, and higher educations than other rural election districts.

This bubble plot shows the percent of households with loans vs monthly average salaries in each election district. The two highlighted in red are Orkhon (lower right) and Umnugovi (upper left). All of the bubbles above 550k average monthly salary except for the two in red are located in Ulaanbaatar.

These two election districts (and the urban workforce where the company offices are) have benefited heavily from the mining industry of Mongolia. Industrialization as proposed in the party platforms would most likely further benefit these areas, as this is where the raw materials would come from.

The second outlier is Bayan-Ulgii. In the chart above, the red bubble to the farthest left is Bayan-Ulgii, and the red bubble on the far right is Khan-Uul District. The average household size is 5.2, significantly larger than the most populace district Songinokhairkhan of 3.8. Besides larger family sizes, Bayan-Ulgii has the lowest average monthly salary.

Ulaanbaatar Districts Have Diverse Needs

The districts of Ulaanbaatar also show diverse needs. Every district also contains part of the ger district. This means that there is a mix of urban and peri-urban in every election district of Ulaanbaatar. In the center of the city, households are connected to the central water, heating, and sewage systems. In the ger district, living standards become sharply different, with no central heat or water connections.

The unique layout of the city districts ensures that at least 20% of the population of each district uses pit toilets and wood or coal for heat. More than 40% of Khan-Uul district households, the most affluent election district in Mongolia, use ordinary wood and coal stoves for heat.

MPs Will Represent Very Diverse Groups

Regardless of the outcome of the election, it is clear that the needs of citizens in election districts are not uniform, and even the common trope of UB vs the rest doesn’t hold up completely. Even though Mongolia a small country with a heavy focus on mining, it isn’t clear that the party platforms address the inequalities mentioned above.

While the economics platforms of the DP, MPP, and Right Person Electorate all focus on heavy industry, it isn’t clear how heavy industry would help rural areas outside of the main mining areas lacking in high paying jobs, or whether it would increase employment levels in Ulaanbaatar.

Mongolian Meat Price Time Series Forecast

Mon, 20 May 2019 08:00:00 +0800

In the land of eternal blue skies and nearly 70 million animals, meat is getting expensive

Since September 2018 the average price per kilogram of mutton in Ulaanbaatar has increased by 35% (beef increased 26%). News outlets have been buzzing with talk of these price increases and even meat shortages. Headlines from many news outlets make it seem as if the sky is falling (see below for a particularly sensational headline from Ikon.mn). I hope to explain the market forces at play here and hopefully better understand why prices are at record highs.

Time series of beef and mutton prices from 2011 to 2019.

How Meat Works in Mongolia

In Mongolia, meat prices are highly seasonal, with meat prices rising in the Spring and then falling in the Fall. The meat supply itself comes from nomadic herders who make up roughly 30% of the population. But these herders do not operate with anything resembling a western mode of production. Their behavior has more to do with herd health and preparations for the various challenges of each season.

In the opinion section of Ikon.mn. Headline translation: “The price of meat is skyrocketing and there is a shortage!”

Here is a rough generalization of the behavior of herders. This, of course, does not speak for all herders but from what I have observed this is relatively typical behavior. In the Fall before winter sets in herders will slaughter or sell those animals which are likely to die in the coming winter. Then in the Spring during breeding season herders often restrict the amount they sell so as not to limit their potential growth that year. In the Summer the productive animals are known and herders are more willing to sell or slaughter those animals that are not pregnant.

With this cycle, it is clear that seasonal prices are primarily being driven by the supply of meat. Prices go down in the Fall, hit the bottom in the Winter, and then rise dramatically in the Spring, reaching the peak in the Summer.

To ensure the supply of meat, the government purchases extra meat when prices are generally low and freeze it for use in times when supply is lower. This program has been going on for several years and the Minister of Food, Agriculture, and Light Industry recently stated they will increase the supply of this reserve meat on the market. However, based on available data it is not clear whether this program actually works or does not. This reserve meat is frozen for several months and, from what I have observed, is much less desirable in the market.

Are increasing exports causing record prices?

Now that we have a short introduction to how the market works in Mongolia, let’s take a look at the factor most are blaming for the price increases, exports. Recently I was interview on MNB World’s weekly news program. Through this, I learned of the concern that international traders (particularly from China) are being blamed for buying meat in bulk, thus causing a meat shortage and prices to rise.

To understand this better I wanted to see if meat prices correlated with exports. After all, meat exports are not a new thing in Mongolia. If these exports were causing a fluctuation in the price of meat we should see these correlations. So let’s take a look!

The blue shaded bar chart is animal product exports (in tons), and the two lines are beef and mutton prices (all prices in this article are Ulaanbaatar averages). As the chart title says, the connection isn’t clear. Export rose dramatically in 2017 and 2018. There may be a lag where 2017 exports impact prices in 2018 and 2018 exports impact prices in 2019, but this doesn’t make too much sense (meat goes bad after a few months being frozen).

In 2017 exports more than doubled from 2016, but average meat prices actually went down. We do see another dramatic increase in exports in 2018, but prices reached the same high from 2015. Simply put, exports may be causing prices to increase, but I do not believe it is the main driver.

Exports may be causing prices to increase, but I do not believe it is the main driver.

3.8% loss in 2018

There wasn’t much talk of dzud last winter. In fact, international NGO’s barely mentioned the fact that last year saw more adult animals die than any year since 2010. A total of 2.6 million adult animals were lost in 2018. Let’s visualize the adult animal loss and meat price.

In 2010 over 10 million adult animals died. It was the last major dzud. Late 2010 then saw a dramatic rise in prices. I’m still unsure why we didn’t see a fast rise in prices. This was the same year Ulaanbaatar saw an explosion of people moving to the city. Perhaps them selling their herds caused a long enough increase in the supply to delay the rising prices.

Nevertheless, the large number of animals lost in 2018 is a very likely explanation for rising prices this Spring. Most likely both are responsible, as well as other external factors like rising salaries or fuel prices.

What happened to inflation?

Since 2010, the consumer price index (CPI) has gone up 84%. Annual inflation is somewhere between 6–9%. Given these numbers, it only makes sense that prices would go up. Yet since 2011 meat prices have been remarkably stable.

To illustrate this I took the monthly change in CPI from MongolBank and calculated what meat prices would be if they tracked inflation.

From 2012 to 2015 both mutton and beef trended above inflation. Then from 2016 to 2018 mutton was below inflation, with beef above. Now we see both again trending above the calculated inflation. What we can learn from this is that rising meat prices are not unexpected.

Prices go up because of inflation. If they don’t go up, but the currency value is decreasing due to inflation, that meat the real price of meat actually went down since 2012 when there was a large spike. This increase in price could simply be a reversion to the inflation trend that you would expect.

Mongolians are richer than they were 8 years ago

Also, let’s remember that since 2010 wages have also increased quite a bit. In 2010 the average monthly wage was 345,000 MNT in Ulaanbaatar. In 2018 it was 1,119,700 MNT. That is an increase of 225%!

Given the currently rising meat prices, do you think that the amount of money the average person spends on meat has gone up since 2011?

If you said yes, think again! Let’s assume that an ‘average’ person in Ulaanbaatar (making an average salary) has a family of three and each person eats 250 grams of meat per day (this is equivalent to the average US meat consumption per person in 2018) for a total of 750 grams of consumption for the family. Based on these assumptions we saw a reduction of around 7% for both mutton and beef as a percent of the average salary spent on meat.

The Tsuivan Index

To better understand the inflation of prices since 2011, I wanted to see how the price of other common items and service has changed. The National Statistics Office of Mongolia (NSO) has recorded the average price of common goods and services going back to December 2010. Below, I’ve created an index for a selection of goods and services by setting the December 2010 price as 100. If prices increase or decrease it will show as a proportion of the December 2010 price (i.e. 100). In this way, you can effectively see the percent change in prices since December 2010.

From the above chart, you can see that beef is about 250% more expensive (2.5 times) than in December 2010. Mutton is about 240% more expensive. What is interesting is the increase in prices for the other items. Here is a description of each:

Restaurant Gulyash — This is the price of a typical goulash meal (including potato salad and tea) in Ulaanbaatar at a typical restaurant.
Tsuivan — This is the price of a plate of Tsuivan (including milk tea) at a ’tsainii gazar’, which is a canteen style of food establishment common across Mongolia.
Men Haircut — The price of a men’s haircut in Ulaanbaatar.
Women’s Haircut — The price for a ‘simple’ women’s haircut in Ulaanbaatar. For obvious reasons, they don’t track more complicated haircuts.

The most shocking price increases were for haircuts. Haircutting is a service business, and there is little in the way of materials needed other than the expertise of the person. In addition, haircuts are not imported and could be said to be a “Mongolian” product.

Of course, haircuts are of course not as necessary as meat for the average person, and haircuts can be performed at home. Other products like chips, Coca Cola, and other packaged foods aren’t tracked by the NSO, but based on my observations since 2012, they have increased significantly in price.

Becoming the Meat Price Prophet

I’m not a big fan of time series forecasts. They are generally too general and not accurate enough to truly understand a problem. Both times I tried previously didn’t give good results. But it seems that for meat prices it might not be too bad! Or is it?

Back in November 2018, I used Facebook’s Open Source Prophet library to forecast mutton and beef prices. For reference (and to show the limitations of time series forecasts), here are the forecasts.

Based on these forecasts our predictions were quite a bit off. But these forecasts don’t take into consideration animal losses, exports, or anything other than the seasonal trends.

Prophet allows the addition of ‘regressors’, which are simply other variables that impact the prediction. What I did this time was take the previous years animal loss and added it to the model. In addition, I took the average yearly loss since 2011 (1.1 million animals) and assumed that would be the animal loss for the next two years. Here is the output of these two models, one for mutton and one for beef. I chose not to forecast other meat types as beef and mutton are by far the most popular.

For each one the blue line is the average forecast, the black dots are the actual average prices, and the blue shaded area is the 80% confidence interval. The short story is that the peak price is almost over and we should expect a reduction in prices in June.

I expect that these forecasts are more accurate than my previous ones because it includes one of the primary drivers of price, animal loss. However, it will be important to add additional features if you wanted to get a more accurate forecast.

Conclusion

Intelligence agents “inspected” meat exporters. I assume they found meat.

Last week the General Intelligence Agency (equivalent to the FBI in the US) conducted a search of meat exporters warehouses. This implies there may be some conspiracy to raise meat prices or that exporters are hoarding meat to then sell back later at higher prices. This seems pretty ridiculous on the face of it.

Meat frozen more than a week is very undesirable on the market in Mongolia, and generally is offered at a much lower price. The Ministry of Food, Agriculture, and Light Industry say that their reserve meat is designed to provide a lower cost option for low-income families. This reserve meat is bought when prices are low (the Fall) and then sold in the Spring. This is not the first choice for any family. In most food markets in Ulaanbaatar, you buy meat that has been slaughtered that day.

Mongolia’s meat prices are a function it’s nomadic supply. Regulating meat prices would only hurt the nearly 30% of the country that is engaged in herding work. These herders also don’t have a large household income compared to Ulaanbaatar. Rising meat prices means more money in their pocket.

There is one question that hasn’t been discussed much. Who is hurt by rising meat prices? Low-income households have government reserve meat as an option. Average and upper-income households (assuming consistent consumption) spend a smaller portion of their income on meat compared to 2010. Let’s not forget that more meat exports contribute to the stability of the Tugrik and help diversify Mongolia’s economy so it isn’t so reliant on mining. Do you want Mongolia to be called Minegolia or Meatgolia?

The Ethical Matrix of Dzud Forecasting

Wed, 13 Feb 2019 08:00:00 +0800

In 2016 a pivotal book in the field of data science was released. Weapons of Math Destruction, by Cathy O’Neil, makes the claim that the algorithms which promise to change our world for the better can also reinforce discrimination, unfair bias, and cause damage on a massive scale.

In this book and in a forthcoming research paper, O’Neil proposes that algorithms don’t just predict the future, they can also create it. As such she proposes the use of an “Ethical Matrix” be completed before such an algorithm is implemented. This matrix is designed to understand the real impact of an algorithm. It aims to explicitly answer who the algorithm is impacting and how in the following ways:

Fairness: is the algorithm truly fair to all those impacted?
Transparency: do we have visibility as to how the algorithm is making its decisions?
False Negatives: how are stakeholders impacted by false negatives?
False Positives: how are stakeholders impacted by false positives?

Why is this matrix important? As algorithms are being applied in more areas of our lives, there is often little regard to their full impact. Here are just a few examples of algorithms and their unintended consequences.

In 2015 Google’s auto-tagging photo feature embarrassed the company by labeling a black person a gorilla. Google still has this problem.
Recently a significant amount of scrutiny was placed on algorithmic credit scoring. These supposedly fair algorithms show a surprising bias in areas such as race. In one example an algorithm reduced credit scores for customers that had a bank charge for “counseling”.
Amazon has used an algorithm to assess recruits applying for jobs at the company. This algorithm appeared to have a strong gender bias against women. Amazon says they never put it in production.

Given these potential consequences, it’s a good idea to evaluate any potential new algorithm to better understand how it could hurt those it impacts. A recently released algorithm in Mongolia inspired me to do just this.

Livestock Mortality Index

In November 2018 a research paper was released by People in Need, an international NGO, that proposes a livestock mortality index to perhaps provide an early warning for dzud in Mongolia. This index would assign a value from 0–5 for each aimag (province) that would describe the risk of dzud in that aimag for the next winter. As dzud is perhaps the largest natural disaster Mongolia faces, being able to accurately assess the risk of dzud is a very important step in making herding a sustainable livelihood long term.

Currently, the Government of Mongolia has a risk map model (at the soum level) in place that was implemented in 2015/2016 and is updated every 10 days. The way in which the proposed index is different is the lead time it would provide to prepare areas that are at the highest risk. Instead of identifying a sudden issue up to 10 days in advance, there would be an indicator of risk months ahead of time. This would give the government and non-governmental actors time to preposition fodder or other materials long before a dzud occurred.

In the paper, the authors make clear that this index is not intended to be a forecast, but rather an indicator of risk. However, as this is essentially an algorithm that is attempting to predict the future, I thought this was the perfect time to apply the Ethical Matrix to a Mongolian case.

Ethical Matrix

To assess the algorithm we will look at four areas: fairness, transparency, false negatives, and false positives. To quickly summarize the algorithm, it uses 7 indicators to create a score from 0–5 (ex. 0.1, 1.4, 2.3, 4, 4.4) for each aimag. The higher the number the higher the risk. The indicators are:

Winter temperature anomaly
Snowfall anomaly
Biomass anomaly
Pasture use anomaly
Zootechnical scores (a combination of past mortality and female reproductivity)

Fairness

The idea of fairness is the general question of whether the algorithm treats those impacted by it equally. In this, the algorithm does quite well. One area of concern is the metric of forage biomass (the amount of forage available). Given the disparity between the south of Mongolia and the north, this is an area where an imbalance could occur. The authors handle this by looking at the anomaly (or change in) biomass to ensure areas with disparate biomass are not treated differently.

Transparency

Transparency in an algorithm is a difficult thing to accomplish. As these are computer models, it isn’t as simple as asking the computer ‘Why did you do that?’. However, what we can do is understand how the inputs are related to the outputs. For example, if I change one thing how does this change the score (higher/lower). In this way, we can understand “how” the algorithm is coming to its conclusions.

The algorithm here again does this quite well. As it is a linear model, we have coefficients for each variable. These coefficients are basically weights on the variables that are combined with the variables themselves to generate a score. The model equation is given and as such we can determine how each variable impacts the final risk score.

Page 16 of the MDVI report

False Positives and Negatives

It’s great to have a fair and transparent algorithm, but what if it fails in certain ways that can cause harm? This is what we try to assess when looking at false positives and false negatives. A false positive in this case is a high-risk score (a “positive” result for high dzud risk) when little to no mortality occurred. The opposite situation is a false negative, where the risk score is low but high mortality results. Both are undesirable but for different reasons.

False Positives

A false positive is also known as “crying wolf”. The result can be similar to the fable, with an erosion of trust in the model after several attempts to prevent disaster over the long term. In the short term, a false positive can waste resources that could otherwise be used in areas that need them more. Over the long term, the larger movement to stop a dzud from happening could be undermined with herders, governments, and NGO’s withdrawing from the program.

The idea of false dzud alarms was discussed in 2017 by Julian Dierkes, a professor at the University of British Columbia and a well-known observer of Mongolia. This critical look at false positives shows how not only data can undermine the system, but the perception of the situation.

False Negative

A false negative can have dire consequences. In this situation, high mortality would occur with the model indicating a low risk. This would mean that enough fodder or other materials may not be readily available on short notice to help in aimags falsely designated as low risk.

This would have more extreme short term consequences not only for livestock but also for trust in the model. If a model has even a few false negatives it very likely could be thrown out and labeled untrustworthy.

Confusion Matrix

In general, when optimizing a machine learning model it is good to ask whether false positives or false negatives are worse. For this model, it is conceivable that several false positives could be tolerated, although with an erosion of trust and possibly less mobilization for each subsequent false positive. However, it would be disastrous to have even a few false negatives in the real world. Livestock could die by the hundreds of thousands and the area could be devastated economically. For this reason, I believe that false negatives for this are much less desirable and should be minimized as much as possible.

The authors of the study made available the predicted risk scores and the livestock mortality numbers for the years in question. With this data, we can create a confusion matrix that will allow us to visualize our false positives and negatives. As the authors did not give bins for the index values, we will simplify things a bit and bin them ourselves. The bins are as follows:

Index, Mortality %
0–1, [0%-3%]
1–2, (3%-6%]
2–3, (6%-17%]
3+, (17%+)

Here is the corresponding confusion matrix. TP = True Positive, TN = True Negative, FP = False Positive, FN = False Negative.

Confusion matrix derived from MVDI risk scores

Viewing this confusion matrix we see a relatively small number of true positives or negatives (about 15% of the total). If this were a classification task this would be quite undesirable. However, the model is set up as a regression task, so this isn’t necessarily bad. Of more interest are the FP’s and FN’s.

The false positives are the large majority here. As we established that we want to minimize FN’s due to their destructive potential, this is a positive sign. However, FP’s are nearly 74% of the values. Given this, it should be known that this model will consistently overstate the risk values.

Conclusion

One way to solve the issue with FP’s and FN’s is to restate the task as a classification one. Currently, as regression, the output plotted against real values is shown in figure 5. While it’s simple to say that higher values mean higher risk, what is one aimag has “1.5” and the other has “2.2”. Should more resources be provided to the higher risk even though we know this model has a high false positive ratio?

Resulting index values from MDVI model

By changing the problem to a classification one we can allow the user of the model to make more intuitive decisions from the output of the model. Instead of numeric values, the output of the model could be “low”, “medium”, and “high”. This may in itself be too simple, but if the classes are based on actual results from real data, then they will have a higher interpretive ability in the real world.

It is an important step for any machine learning or algorithmic project to take time to understand how it’s results can impact stakeholders and the system as a whole. The authors state that this model was not designed as a dzud forecast. However, once they release it in the real world NGO’s and governments can interpret the results how they wish. As such, there is a very real responsibility to understand how these values can be interpreted.

The Tsagaan Sar Debt Cycle

Fri, 08 Feb 2019 08:00:00 +0800

The entire country mobilizes weeks ahead of time to prepare. The city government, worried about traffic, only allows cars to drive during alternate days on the weekend. The bakeries are so consumed with making ul boov, a traditional pastry, that bread starts to disappear from store shelves the week before the holiday.

This sounds like quite a lot, but it is just the run-up to the actual event, Tsagaan Sar, Mongolia’s lunar new year celebration. The main activity of the holiday (which goes on for three days) is where family (and extended family) visit their elders in their home. The guests traditionally bring money as a gift for their elders, and the guests are given gifts (sometimes small, sometimes large) in addition to a truly magnificent feast.

The video below shows the preparation of one of the signature dishes of Tsagaan Sar, buuz.

Tsagaan Sar Preparation

The Cost

Each year new outlets write many articles sharing the costs of Tsagaan Sar. News.mn estimated this year that cost for one family is about 1.32 million MNT (about $500 USD). To put his in perspective, more than half of those employed make less than $200 per month. A survey completed by a market research firm and reported by BloombergTV Mongolia concluded that 62% of those surveyed take out loans for the holiday.

I’ve thought about the cost of Tsagaan Sar for a few years now, but not in absolute costs. I’ve written before about the growing debt in Mongolia, and I wanted to see how the holiday impacted those who hold the greatest burden, the elders responsible for hosting 10’s or 100’s of guests. Phrased in a testable way:

Do pension or salary loans issuances show an increase for Tsagaan Sar separate from the rest of the year?

Loan Data

Luckily, MongolBank, the country’s central bank, keeps statistics many statistics on loans. I chose salary and pension loans because they are issued to individuals and are not (directly) used to purchase real estate. Pension loans are of particular interest because they are given using the government pension of the person as collateral.

To do the analysis I downloaded the quarterly report of " Individual and SME loans issued by banks" and separated out salary and pension loans issued. The data is only available quarterly or annually. Quarterly actually suits our needs quite well as Tsagaan Sar always takes place in Q1 while the exact dates move every year.

The head of our issued loans data. All numbers are in millions of MNT.

Let’s first take a look at the time series of both salary and pension loans.

Other than the obvious large increases in both, the thing to notice is when each increases during the year. It is already noticeable that pension loans have a spike at the beginning of each year while salary loans spike later.

To do a trend analysis we can use the handy Prophet package from Facebook. There is a built-in seasonality tool that is perfect for this analysis. The tool is built on daily data analysis, so changes were made to accommodate a quarterly period (mainly reducing the Fourier order for seasonality). The graphs below show the derived trends.

It is important to remember when interpreting these graphs that as we only have one data point per quarter, Prophet is interpolating the trends between the quarters. For reference, quarters end March 31, June 30, September 30, and December 31.

Salary loans do show a large trend in Q1, but the largest trend appears to be in Q3. Perhaps parents trying to get enough money to pay for school tuition, clothes, etc.?

Pension loans have an obvious upward trend in Q1, with Q2 and Q3 being significantly smaller. Can we answer our question?

Do pension or salary loans issuances show an increase for Tsagaan Sar separate from the rest of the year?

Well, we certainly see a large increase in pension loans in Q1, which is when you would expect Tsagaan Sar expenditures, and therefore the need for cash. This certainly seems to fit with the narrative that the elderly are the most impacted, and are taking out loans specifically for the holiday.

Perhaps if you visit a family next year for Tsagaan Sar you may consider giving a bit more money to help offset these costs.

Mongolia's Economic Future is Chained to China

Mon, 21 May 2018 08:00:00 +0800

That is, unless Mongolia’s export partners diversify

It’s a rather gloomy title, but it is appropriate. There has been much discussion about diversifying the Mongolian economy to avoid the boom and bust cycle that has become common. Less has been discussed regarding diversifying trading partners.

For those familiar with the Mongolian economy, it is clear that China plays a big role in imports to the country. Yet I believe it is less known by the average Mongolian how closely exports are tied to China as well.

With all of the efforts of diversification, is the economy safer if nearly all of Mongolia’s exports go to one trading partner? Now factor in Chinese government whims to close the border whenever it fancys. This seems to be a recipe for economic uncertainty.

To understand just how closely Mongolia’s exports are tied to China I gathered data from the National Statistical Office and dug in. These data show Mongolia’s export value from 2005–2017 and is broken down by country.

Mapping Exports

First I wanted to visually see where Mongolia’s products and services are going around the world. I animated a global map, showing a new image for each year, and shaded the countries by the amount of exports going to that country. The darker the higher the value.

China remains dark throughout the period while other trading partners get significantly lighter. There is only one country that ends darker than it started, the United Kingdom. Interestingly enough back in 2005 the United States was number two, yet was still far away from the export value of China.

Charting Exports

Let’s take a look at real numbers now. But I warn you, they aren’t any prettier.

Here you can see the massive proportion of exports going to China from Mongolia. Every other country is dwarfed.

Total Exports

Of course it’s not all doom and gloom. Since 2005 Mongolia’s exports have increased tremendously, propelling the country’s economy to new heights. The total value has gone from just over 1 billion USD to over 6 billion USD today.

Total value of Mongolian exports with a linear forecast. Error bars shown are based on a 95% confidence interval.

Mongolian exports will continue to grow, and hopefully in a diversified way that includes more than natural resources and more than just China. Putting all your eggs in one basket is never a good idea.

Mongolians are more in debt now than ever

Tue, 08 May 2018 08:00:00 +0800

But they are also richer.

The economy is picking up steam in Mongolia. This is good news for the country’s businesses, who can hope the economic growth will translate to more money in peoples pockets. However, in a country that has experienced dramatic up and down swings of the economy, hope may not be in large supply.

As I always tell my students, if you have a question, dig into it and test your hypothesis. My hypothesis is that debt levels may be reaching a critical amount that the economy can no longer bear. This will result in a lack of free cash for regular citizens and a weakening of non-mining economic growth. But hey, don’t take my word for it. Let’s look at the data.

A huge amount of construction has taken place in recent years along with an increase in average salaries. Financing all of this building are commercial loans and also a government backed mortgage program offering the lowest rates in the country. This has led to an explosion of outstanding loans to finance the building of apartments and businesses.

Below I’ve charted outstanding loans and deposits from 2001–2016 (the most recent year available from the national statistical office). After the mining boom in the early 2010’s loans skyrocketed.

The amount of outstanding loans is nearly 12.5 trillion MNT (just over 5 billion USD). At the same time deposits have grown significantly, but not at the same pace as loans. Ulaanbaatar leads the country in outstanding loans with 78% of all outstanding loans. These numbers combined with estimates in 2017 that there are 35–40 thousand empty apartments in Ulaanbaatar are quite concerning.

Is this debt trend national? Below I mapped the total of outstanding loans per capita in 2016 for each province (aimag) using Tableau. I excluded Ulaanbaatar as it highly imbalanced the map. As this is per capita we are balancing each provinces population to better compare them. For comparison Ulaanbaatar had a total outstanding loans per capita of 26.88 million MNT (nearly 11,000 USD).

It seems pretty clear debt is consolidated primarily in Ulaanbaatar. This is both good and bad. Nearly half the population and most of the economic activity of the country is in Ulaanbaatar. However this also probably means the average UB’ier has more debt than the average person in the countryside. With youth unemployment over 20%, debt in the nations capital may prove too much to bear.

A first look at Mongolian crime data

Thu, 30 Nov 2017 08:00:00 +0800

Note from April 2021: The crimemap referenced in this no longer works. It seems the government of Mongolia is no longer supporting this project.

A few months ago I stumbled across a very interesting site made by the Mongolian national police that allows you to map crimes all around Mongolia. At the same time I was introduced to Google Data Studio, a relatively new product from Google that allows you to create dashboards from live data. I thought it was a natural fit to mash the two together and make an attempt at a “Mongolian Crime Dashboard”. The rest of the article I will explain about this dataset, its features, and how I mashed it up with Data Studio.

What the Mongolian Crime Map looks like. Looks amazing but the locations are suspect. Source: crimemap.police.gov.mn

How the Mongolian Crime Map Is Different

At first glance the Mongolian crime map was similar to crime maps of Chicago. You can visualize categories of crime over a specified time and it shows the locations of each crime. But the Mongolian data are less complete and messier than what would be ideal. The fields include the crime type, report number, date of report, location data down to the khoroo (neighborhood), and police station. However a few differences are clear when I dug in deeper.

First of all the location data used to map the crimes on the map do not appear to be entirely reliable. Crimes appear to be mapped outside of their areas in some cases. Some crimes appear in entirely different countries. For example I found two crimes in Europe, a few in Siberia, and a few in China. So I determined using the mapped coordinates was a less than perfect way to use this data.

Second is the report location. It is a text field with the city/province, soum (like a county), district, and khoroo (sub-district / neighborhood). All of these details are separated by commas or spaces, with quite a bit of variability. I was able to use some regex functions in good old Excel to clean it up. Not ideal but workable.

Last is the time period available. Only 2015 and 2016 appeared to be complete. 2014 and 2017 seem to be incomplete as the total number of crimes is quite a bit lower. As such I focused on 2015–2016 for the dashboard. My theory as to the incomplete data for 2017 is either that they haven’t entered the reports or they only enter cases that are closed after a certain period of time. Either way it should be understood that there may be some missing data in the years 2015–2016 as well.

Mongolia’s National Statistical Office

Mongolia’s National Statistical Office (NSO) does a great job of collecting crime totals as well. However their data are only reported down to the soum/county (outside Ulaanbaatar) or district level (inside Ulaanbaatar). This makes comparing neighborhoods difficult as some districts are quite large. Nearly 50 kilometers long in some cases.

Map of Ulaanbaatar districts. This is the smallest unit the NSO records crime data for. Credit: The Mongolist

A major advantage is the NSO has complete data from 2010–2015, much longer than what is available from the Mongolian Crime Map. This historical data could be very valuable in things like crime forecasting, which could be very helpful in policing. However this historical data is compiled yearly so we would be unable to see trends outside of year to year.

The Dashboard

Google has come a long way since launching Data Studio last year, and the product is still in beta (per usual with Google). It is already a very powerful tool and clear competitor to Microsoft’s Power BI. Even things like geo maps are relatively simple to make and are well documented. These are normally challenging to make as most software is geared to the developed world.

Mongolian crime dashboard. Google Data Studio is limited to mapping down to the province level.

Time series look great and are easily filterable, although are a bit messy.

Comparing Ulaanbaatar crime to provincial crime for Q4 2016.

It is important to note that much of the labels are in Mongolian, but Google Translate does a pretty good job on most of the labels if needed.

Have suggestions or comments? Feel free to drop a comment below.