<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Amogh Arakali on Medium]]></title>
        <description><![CDATA[Stories by Amogh Arakali on Medium]]></description>
        <link>https://medium.com/@amogharakali?source=rss-17a040da3940------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*wZ8Jv8b330Qt3OAKmdOt3w.jpeg</url>
            <title>Stories by Amogh Arakali on Medium</title>
            <link>https://medium.com/@amogharakali?source=rss-17a040da3940------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 05:37:51 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@amogharakali/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Applying to a PhD (Part 2 — Costs)]]></title>
            <link>https://amogharakali.medium.com/applying-to-a-phd-part-2-costs-78ff0f5d3352?source=rss-17a040da3940------2</link>
            <guid isPermaLink="false">https://medium.com/p/78ff0f5d3352</guid>
            <category><![CDATA[education-costs]]></category>
            <category><![CDATA[doctorate]]></category>
            <category><![CDATA[phd-application]]></category>
            <category><![CDATA[phd]]></category>
            <dc:creator><![CDATA[Amogh Arakali]]></dc:creator>
            <pubDate>Mon, 11 Sep 2023 22:01:47 GMT</pubDate>
            <atom:updated>2023-09-11T22:35:53.993Z</atom:updated>
            <content:encoded><![CDATA[<h3>Applying to a PhD (Part 2 — Costs)</h3><h4>Applying to US programmes for Indian students</h4><p><strong><em>Note: </em></strong><em>My post here is largely meant for Indians who are considering applying to social sciences or interdisciplinary PhDs in the United States (with some relevance for Canada). Other aspirants may also find this post useful, but please be aware that I’ve written this post with the above audience in mind.</em></p><p><em>Please also note that this is based on my experience with the Application Process. I applied mostly to Planning, Public Policy, and Urban Studies PhD programmes. Other people may have had different experiences. Speak to a wide number of applicants to get different perspectives.</em></p><p><em>Disclaimer — I am not a professional financial advisor. Therefore, do not assume this post to be professional financial or investment advice. This is just learning from personal experience.</em></p><p>— — — — —</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*J_h2B_sMtpUgqmHBFs37GA.png" /><figcaption>AI Art, made using Dall-E. Prompt: “Black silhouette of a university student in front of a screen depicting costs of tuition, with the white silhouette of the university campus against a blue sky behind them.”</figcaption></figure><p>This is the second of my three posts about applying to a PhD. This post is largely about costs. Please note that these are the costs involved in <em>applying </em>to a US PhD<em>, not the cost of the PhD programme itself.</em></p><p>I’ve listed the major costs of applying below. Please note that while some of them are needed only after admission, it’s best to begin planning for them around the same time you begin your application. Some of them can take months to arrange.</p><p>(My) major costs during the Application Process were:</p><ol><li>Application Fees</li><li>GRE and TOEFL Costs</li><li>Transcript Costs</li><li>I-20 Costs (after admission, but prepare for this at least 6–7 months before admission results come in)</li><li>US Visa and SEVIS Fees (after admission)</li><li>Deposits (after admission, but prepare for this beforehand)</li><li>Flight Tickets (after admission)</li></ol><p>I’ve written about each of these costs in more detail below.</p><p>— — — — —</p><p><strong>APPLICATION FEES</strong></p><p>I’ve already mentioned these in my previous post, so I won’t talk too much about them here. However, keep in mind that application fees will probably be one of the largest components of your expenses. Each application will cost between $80 to $120, so the more universities you apply to, the higher your expenses will be.</p><p>That being said, I wouldn’t recommend applying to too few universities, unless your finances don’t give you any other choice (this is a strictly personal opinion — others may disagree with me, please ask around). As I mentioned in my previous post, rejection rates for PhD programmes are very high. If you’re making the effort to apply to 2–3 universities, you may as well apply to 6–7 instead, and increase your probabilities of acceptance. Your application costs will be high either way, but by applying to 6–7, your chances of getting in will be higher as well.</p><p>— — —</p><p><strong>GRE AND TOEFL COSTS</strong></p><p>Let me get something out of the way. Since the pandemic, many US universities and departments have dropped the GRE requirement for PhD applicants. I don’t know if GRE will come back as a requirement in the future, but for now, most applications are GRE-free.</p><p>My (strictly personal) opinion is that if the majority of the universities in your list do not require the GRE, then <em>do not write the GRE.</em> Just drop the universities which do require them. There’s no value in spending Rs. 30,000 to write an exam that will only be required by 1–2 universities in your list (assuming you’re applying to 6–7 places).</p><p>I took the GRE and I consider it a mistake. There was one university (one of my top choices) which required the GRE. Since it was a top choice, I registered for the exam, spent three months preparing for it, got my scores, gave the application my best…and got rejected anyway. For a single university, this was a waste of Rs. 30,000, a sum I had spent almost a year saving up for. Maybe the GRE may come into use later on (some people say GRE scores help with scholarships, but I’m not sure about this). As of now, I consider it money ill-spent.</p><p>On the other hand, TOEFL may seem unnecessary, but if you’re applying to 6–7 places, it’s likely that they will all require you to have TOEFL scores. In my opinion, paying the money, writing TOEFL, and submitting your scores is a lot easier than writing to 6–7 different places, showing proofs that you’ve studied in English-speaking institutions, and asking each department for a TOEFL waiver (one of my best friends will disagree with this, but still :D). If you’ve studied in a good English-medium school or college in India, TOEFL doesn’t require more than a week of preparation, so it’s relatively effortless.</p><p>TOEFL is still expensive though, about Rs. 17,000. If you can’t afford it, you may still need to write and ask for waivers. Keep in mind that GRE and TOEFL are both run by the same organisation (Education Testing Service or ETS). They often give you discounts if you apply for both together. I got to write TOEFL for free because I paid for GRE (still, I don’t think the GRE was worth it).</p><p>— — —</p><p><strong>TRANSCRIPT COSTS</strong></p><p><strong>Note:</strong> Most US universities will not require you to submit official transcripts <em>until you’ve been accepted into their programme</em>. You can submit “unofficial transcripts” during your application and submit official ones after admission. What this means is that during the application process, they’re usually willing to accept a scan or PDF of your actual transcript, as long as it carries official letterheads, signatures, contact details, and seals. Check with the US university about this.</p><p>Nevertheless, for me, <strong><em>sending my official transcripts to the US was, without any debate, the worst part of my application experience.</em></strong> A big reason for this is that there was very little information available on what to do. So I made quite a few mistakes along the way before I set things right.</p><p>However, I should add that how bad or how good the transcripts experience can be, depends entirely on your US and Indian universities, the systems they have in place, and what information is available. I know friends who also got into PhDs, who had no issues with transcripts at all. So check up with your US and Indian universities in advance, and find out about processes for sending transcripts across. It may save you a lot of trouble later on.</p><p>Cost-wise, I had five headings to consider:</p><ol><li>The cost of getting transcripts issued from my undergraduate college.</li><li>The cost of getting transcripts issued from my post-graduate institution.</li><li>The cost of getting (1) and (2) signed by the Registrar of my University and sent to the US.</li><li>The cost of getting my signed transcripts verified by an authorised US intermediary (more on this below).</li><li>The cost of co-ordinating, logistics, travel, and stay while this was going on.</li></ol><p>US universities have very specific requirements on how your transcripts should be sent to them. A general principle is that you, the applicant, cannot handle the final transcripts that go to the US. <em>These must be sent to the US directly from the Registrar’s offices of the Indian universities you’ve studied under</em>. This means you will have to follow whatever process your Indian university has for getting transcripts sent from the Registrar’s office. And please note, <em>this is almost always the Registrar,</em> not the Principal of your college, nor a Professor in your Indian university.</p><p>Some US universities have tighter requirements. Even after transcripts are sent out from the Registrar’s office, they may not accept transcripts directly. Instead, <em>they may require an authorised private agency (like </em><a href="https://www.wes.org/"><em>World Education Services</em></a><em>) to verify that your Indian university is authentic, and that the documents your Registrar has sent out are genuine.</em> Sometimes, you’ll have to pay for the costs of this verification as well. In some other circumstances, they may require the scores from your transcripts converted into equivalent US grades or GPAs. The authorised intermediary may do this as well, for a price.</p><p>Finally, if you’re like me, your Indian university may be based in a different city from where you’re living currently, and you may be required to travel to your Indian university to get everything done. I made quite a few blunders here. I realise now that I need to have travelled only once. But since there was little clear information, I made two trips to Mumbai from Bengaluru, and nearly booked a third.</p><p>Nevertheless, there was a point when I was co-ordinating between five institutions (my undergrad college, my postgraduate institution, my Indian University’s Registrar’s Office, the US intermediary, and my US university) across three cities (Bengaluru, Mumbai, and New York) and two countries (India and the US). It was maddening and expensive.</p><p>To be clear, <em>everyone at these five institutions was helpful</em>. There were no fights or arguments or unnecessary drama— it was only the process of co-ordination and the lack of clarity that was frustrating. So please make sure you are aware of the processes involved in sending your transcripts to the US before you begin spending on this.</p><p>For me, (4) and (5) above were the most expensive parts of the process, costing several thousand rupees. (1), (2) and (3) were reasonably okay cost-wise, though many of logistical costs in (5) came from lack of information about these three.</p><p>— — —</p><p><strong>I-20 COSTS (PROOFS OF FINANCIAL CAPACITY)</strong></p><p>I-20 costs aren’t really direct costs, but they still require you to have a significant amount of money with you at the right time. So it’s necessary to plan for them well in advance, preferably around September/October the previous year (if not earlier).</p><p>What is I-20? A visa grants you entry into the United States, but what allows you to stay and study there is something called an <a href="https://studyinthestates.dhs.gov/students/prepare/students-and-the-form-i-20">I-20 form</a>. It will carry your details and specify how long you’ve been admitted for.</p><p><strong>This part is really important.</strong> Your I-20 form will be issued by your University’s Designated Student Officer or DSO (usually an authorised official at your University’s International Students’ Office). However, before the DSO grants you an I-20, <strong><em>they will usually ask you for proof that you can afford to stay in the US for the duration of the programme.</em></strong></p><p>This means that <em>at the time of accepting your offer, you will need to have proof of enough money to afford paying for the programme and staying in the US. </em>This includes proof of scholarships, verified commitments from your sponsors (like your relatives), and of course, money you’ve saved in your bank account.</p><p>Income Tax statements or statements of stocks and mutual funds are <em>not considered proof</em>. What usually work are savings accounts or Fixed Deposit statements in your name (or your verified sponsor’s name), and official letters from scholarship-granting authorities. Note — if you’re submitting Fixed Deposits as proof, make sure the FDs are maturing before the start of your academic year.</p><p>Your I-20 can take up to a month or more to be issued, so make sure you submit your proofs and other requirements as soon as you accept the offer. Keep these ready and available by March-end. Make sure that your bank statements do not have large sums of money moving out within the last year. Your DSO needs proof that you’re financially stable and capable of affording the programme.</p><p><strong><em>You cannot apply for a US Student Visa until your DSO sends you your I-20.</em></strong></p><p>On a side note, do remember that your I-20 is as important as your US Visa, if not more so. You will be required to carry it with you when you board your flight, and you will be asked to submit it whenever you deal with any bureaucracies in the US.</p><p>The I-20 also has an expiry date — if your I-20 expires, a countdown for you to leave the country will begin. If your I-20 expires when you’re back home in India, you won’t be allowed back into the US, even if your visa is still valid.</p><blockquote><strong><em>WHAT THIS MEANS PRACTICALLY</em></strong></blockquote><blockquote>For your I-20 application, keep aside <strong>proofs of enough funds for at least 14-18 months of living</strong>*<strong> </strong>in the United States, including the costs of paying for your programme. These do not entirely have to be your own savings, but it should be money formally dedicated to you (scholarships, funds by sponsors, etc.)</blockquote><blockquote><strong>Note that “scholarships” include any scholarships that the University itself has offered you, so if you get a full tuition scholarship or a 75% scholarship from the University itself, then you don’t need too much in your bank account or sponsor’s funds.</strong></blockquote><blockquote>You will need to have these proofs ready at the time of accepting your university’s offer. So begin planning for your proofs at least 6–7 months before, if not longer. Your university will usually require you to accept an admission offer in April. So start preparing your proofs no later than September or October the previous year.</blockquote><p>*<em>or however long your DSO requires you to show funds for. If you’re uncertain about the exact funds you need proof for, write to the International Student Offices of the universities you’re planning to apply to. Ask them how much funds are international students in your programme typically required to demonstrate proofs for, to obtain an I-20.</em></p><p>— — —</p><p><strong>VISA and SEVIS Fee</strong></p><p>Once you get your I-20, you’ll need to apply for a US Visa (usually an F1 visa). There are three kinds of costs you’ll need to plan for here. Two of these are:</p><p>(1) <a href="https://in.usembassy.gov/visas/">The fee for the US Visa itself</a><br>(2) <a href="https://www.ice.gov/sevis/i901/faq">The SEVIS Fee</a></p><p>At the time of your Visa interview/document submission, you will need to have paid for both the Visa and for SEVIS and have proofs of the same.</p><p>The third cost may vary depending on where you live. Remember that US consulates operate in only five Indian cities at the moment — Delhi, Mumbai, Chennai, Kolkata, and Hyderabad. If you live outside these cities, you will need to plan for costs of travel and accommodation for your visa interview. You may also need to account for the costs of using an authorised intermediary like VFS Global.</p><blockquote><strong><em>WHAT THIS MEANS PRACTICALLY</em></strong></blockquote><blockquote>Check the US Visa and SEVIS sites for fee payment processes and make sure you’re aware of how much they cost. Budget for this early, as soon as you send your acceptance to your university.</blockquote><blockquote>Keep track of train/bus/flight costs to the US Consulate you’re applying to, as well as hotel charges.<strong> Keep in mind that many students will be applying for US Visas in June-August, so everything from available interview slots to available hotel rooms will be booked up quickly in this period.</strong></blockquote><blockquote>If you need to use intermediaries like VFS Global, make sure you’re aware of how much they charge, early on. I didn’t need to — I carried out the visa process on my own, but someone else might have a greater need.</blockquote><p>— — —</p><p><strong>DEPOSITS</strong></p><p>In addition to the costs above, you will also need to have money in hand to put down deposits to lock certain things in place. I had to pay two significant deposits:</p><ol><li>A deposit to confirm my acceptance of my admission offer to my university, in April.</li><li>A deposit to pay for my hostel accommodation, in May. For other people, this may be a rental deposit at an apartment, for which you may have more time (depending on location)</li></ol><p>You will be required to send these deposits between April and August.</p><p>These deposit requirements will vary from university to university. Some of them may not require these deposits or have other kinds of deposit requirements. But they are likely to be significant amounts. Overall, I suggest saving at least Rs. 80,000 for deposits, or a lakh if you can afford it.</p><p>The good thing about these deposits is that you get your money back (or they will get deducted from other expenses in the future). Nevertheless, you will need to have this much money at hand to make the deposit at the right time.</p><p>— — —</p><p><strong>FLIGHT TICKETS</strong></p><p>Last, but not least, budget for your flight tickets. The earlier you book, the cheaper they will be. Some airlines also have special offers for student travellers if you can show them your Student ID or your Student Visa.</p><p>My advice is to book your flight tickets as soon as you get your US Visa. Some people book it even earlier, but there’s a risk of losing your money if your visa process is delayed and you have to cancel your flights. If you’re booking before your visa arrives, look for refundable tickets, not non-refundable ones.</p><p>Typically, most students fly out to the US in late August, as the academic year begins on 1 September. By early June, ticket prices for late August had started touching a lakh or more. So I would really suggest booking as early as you can.</p><p>— — — — — — — — — — — — — — — — — — — — — — —</p><p><strong>FINAL NOTE FOR PART 2</strong></p><p>This may all sound intimidating. However, please understand that my objective here is not to discourage you from applying, but to ensure that your application process is as predictable and smooth as possible. Knowing about these costs beforehand and planning for them in advance matters a lot. If you still feel these costs are too much, then take some time and rethink whether you want to apply this year.</p><p>If you do choose to apply in 2023 (for Academic Year 2024), then now (September 2023) is the time to look at your finances and begin preparing for these costs (if you haven’t done so already). Don’t wait until you start submitting forms to begin thinking about these things.</p><p>Most importantly, remember that even after you choose to start your applications, there will be moments during the process you’ll be tempted to give up the process, because the costs seem so intimidating. <strong><em>At these moments: Stop, Take a Deep Breath, and Walk Away for a Few Minutes.</em></strong> Remind yourself that you’ve made these calculations before and that you found ways to make the money and accounting work.</p><p>Come back with a cooler head and look at those finances again. If you’re still feeling intimidated, talk to someone with more financial expertise than you. A lot of the process of dealing with costs is psychological. This doesn’t mean it’s unimportant. Quite the contrary. It’s important that you’re psychologically capable of dealing with money leaving your account (as I mentioned in Post 1). Don’t forget — if you get into a good programme, in the long run, you’ll make that money back. Until then, it’s a question of managing cash flows.</p><p><em>Note about AI Art:</em> I sometimes use AI art in my blogposts, strictly for content I put out for free, where I wouldn’t have hired an artist anyway. I use free AI platforms, or free options, and share prompts/sources so that anyone else can create AI art. Whenever I have time, I draw instead. I <em>do not</em> support the use of AI art when it displaces human artists.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=78ff0f5d3352" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Applying to a PhD (Part 1 — Mindsets)]]></title>
            <link>https://amogharakali.medium.com/applying-to-a-phd-part-1-mindsets-b25e8f694f0f?source=rss-17a040da3940------2</link>
            <guid isPermaLink="false">https://medium.com/p/b25e8f694f0f</guid>
            <category><![CDATA[academia]]></category>
            <category><![CDATA[phd-application]]></category>
            <category><![CDATA[doctorate]]></category>
            <category><![CDATA[phd]]></category>
            <category><![CDATA[doctoral-research]]></category>
            <dc:creator><![CDATA[Amogh Arakali]]></dc:creator>
            <pubDate>Sun, 10 Sep 2023 22:01:42 GMT</pubDate>
            <atom:updated>2023-09-11T10:19:06.611Z</atom:updated>
            <content:encoded><![CDATA[<h3>Applying to a PhD (Part 1 — Mindsets)</h3><h4>My Suggestions for Applying to US PhD Programmes as an Indian Student</h4><p><strong><em>Note: </em></strong><em>My post here is largely meant for Indians who are considering applying to social sciences or interdisciplinary PhDs in the United States (with some relevance for Canada). Other aspirants may also find this post useful, but please be aware that I’ve written this post with the above audience in mind.</em></p><p><em>Please also note that this is based on </em><strong><em>my </em></strong><em>experience with the Application Process. I applied mostly to Planning, Public Policy, and Urban Studies PhD programmes. Other people may have had different experiences. Speak to a wide number of applicants to get different perspectives.</em></p><p>— — —</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1zWBVUYWgk7jPu76a43Qfg.png" /><figcaption>AI art, created via NightCafe. Prompt: “PhD applicant at a university library computer”</figcaption></figure><p>In March 2023, I concluded nearly 12 years of work at the <a href="https://iihs.co.in/">Indian Institute for Human Settlements</a> in Bengaluru. I began preparing to move to the United States to commence pursuing a <a href="https://www.newschool.edu/milano/public-urban-policy-phd/">PhD in Public and Urban Policy</a> at the <a href="https://www.newschool.edu/milano/">Milano School of Public Policy</a>, one of the Schools of Public Engagement at <a href="https://www.newschool.edu/">The New School</a> in New York.</p><p>I’ve written about my motivations for pursuing a PhD <a href="https://amogharakali.medium.com/life-and-career-update-move-to-the-united-states-7f57b2aef88e">elsewhere</a> — that’s not the objective of this post. This post (and the next couple) are about the Application Process itself.</p><p>Many other PhD applicants and scholars have written about planning and preparing for the Application Process and this is my addition to that literature. I hope what I write below will prove useful to people who are looking to apply for a PhD themselves.</p><p>— — —</p><p>This will be the first of three posts. The first post will deal with mindsets and approaches I found useful while preparing my application. The second will deal with costs. The third will deal with logistics.</p><p>This first part is a bit ruminative. It’s deliberately so. I think it’s important that aspiring applicants are aware of abstract things to keep in mind before they apply. Therefore, I’ll talk about mindsets which were greatly useful to me during the Application Process. These mindsets helped keep me sane, and didn’t let me falter along the way. While I taught myself some of this, much more came from other scholars and colleagues who advised me during the Application (shout-out to them for all their help).</p><p>If you’re applying for a PhD, I recommend you read the full text below carefully. However, if you just want to start with some practical advice before moving on to the philosophical, I recommend starting with the sections titled “What This Means Practically”.</p><p>— — —</p><p><strong>Hidden Curriculum</strong></p><p>Before I talk about mindsets, I should share a term which will be useful for both parts of this post — <em>Hidden Curriculum</em>. Hidden Curriculum refers to all the requirements, tasks, work, or behaviours that students are expected to know or perform at university, but which are never explicitly shared, spelt out, or mentioned.</p><p>Hidden Curriculum topics are often things which university administrators or professors take for granted. They assume that all students are aware of such requirements (when in reality, they aren’t). On some rare occasions, Hidden Curricula may be open secrets, which are not talked about for a variety of reasons.</p><p>Application Processes can be full of Hidden Curricula. They will vary from country to country and university to university, but pretty much every place have them.</p><p>There are two points potential applicants need to be aware of regarding Hidden Curricula:</p><ol><li>First, <em>do not assume that Hidden Curricula are caused by malice</em> (<a href="https://en.wikipedia.org/wiki/Hanlon%27s_razor">Hanlon’s Razor</a> is a useful tool here). There are often strong reasons why they exist, although these reasons may not work in your favour. There’s no point getting angry or upset about them during the Application. They exist. We have to deal with them.</li><li>Second, <em>always assume that Application Processes will contain hidden curricula</em>. In other words, always make sure you have extra time, extra energy reserves, and the mind-space to deal with unexpected tasks or unplanned work. Trust me, there will be unexpected stuff.</li></ol><blockquote><strong><em>WHAT THIS MEANS PRACTICALLY</em></strong></blockquote><blockquote>The best way to deal with Hidden Curricula is to give yourself more time. <em>The earlier you start the Application Process, the better it is for you.</em></blockquote><p>— — —</p><p><strong>MINDSETS</strong></p><p>Okay so, on to the mindsets. I’ve found the following points very useful in dealing with the PhD Application Process.</p><p><strong><em>(1) Be prepared for a high rejection rate:</em></strong> <br>Unless you’re an extraordinary student, you’re likely to see most universities reject you.</p><p>This has nothing to do with you. I’ll say it again — this has <em>nothing</em> to do with you. Most universities have the budgets and staff to support 5–6 PhD entrants every year (sometimes fewer). If you’re lucky, your preferred university may admit 9–10 applicants. At the other end of the Application Process, they can get anywhere between 100–500 reasonably good applicants every year. Maybe more.</p><p>Each of these applicants may be a perfectly good candidate, but with such tight acceptance rates, <em>the question of who gets in and who doesn’t can boil down to razor-thin slices of luck</em>. Maybe someone in the faculty has found funding to start a new research theme in the department (which hasn’t even been publicly announced). By sheer coincidence, your application profile fit the requirements of the new theme. You get accepted. On the other hand, another university may be realigning their long-term plans after decades. Your profile (which would have been a great fit two years ago) no longer matches their priorities. You get rejected.</p><p>These things are mostly out of your control. More importantly, <em>this means that rejections are not judgements on your abilities or skills as a researcher</em>. <strong>Do not let anyone tell you that a PhD application rejection is somehow a black mark on your career record. </strong>It’s not.</p><blockquote><strong><em>WHAT THIS MEANS PRACTICALLY</em></strong></blockquote><blockquote>A high rejection rate means you may have to attempt getting into a PhD programme two or three times. I hope this won’t happen, but this is possible.</blockquote><blockquote>Mentally, prepare for at least two years (two cycles) of attempts. I would also suggest giving a gap of one or two years between attempts.</blockquote><blockquote>Giving yourself a year in-between creates some time where you can reflect on your previous attempt and prepare differently for the next one.</blockquote><blockquote>I wouldn’t recommend rushing into the next cycle without some reflection and time off.</blockquote><p><strong><em>(2) Be prepared to spend money:<br></em></strong>This is a Hidden Curriculum topic that very few people talk about (in my experience). The <em>Application Process itself</em> can be quite expensive. The actual expense depends on the country you’re applying to, and the universities you’re engaging with, but regardless, you will need to start setting money aside for the application process early on.</p><p>Example: Most universities in the United States have an application fee ranging from $80 to $120. At a Rupee-Dollar exchange rate of Rs. 82 to $1, that’s about Rs. 7000 to Rs. 10,000 per application. If you’re applying to six or seven universities, the total cost of submitting the application alone can be Rs. 42,000 to Rs. 70,000.</p><p>This does not include the cost of writing the TOEFL (which most US universities require for Indian students), the GRE (which some universities still ask for), or the cost of getting your transcripts transmitted out of India and verified by a US agency (which is a costly logistical nightmare that I’ll talk about elsewhere).</p><p>There are further costs to pay after getting an offer letter, which is again something no one talks about. You may have to pay a deposit to confirm that you’re accepting an offer (you’ll get the money back, but you need to have it in hand at the right time). You may also have to pay a deposit for the housing or accommodation you’ll be staying in. And of course, there are costs to getting a visa and booking your flight tickets.</p><p>This may all sound rather gloomy, but…<strong>I’m not saying any of this to discourage you! </strong>Quite the opposite. I’m warning you about this beforehand, precisely so that you can plan and prepare for these costs when you apply.</p><p>It’s easy to get discouraged when these costs keep popping up without warning and you may give up your application half-way. Knowing about them beforehand can help you plan your finances early on and avoid any psychological fallout from the process.</p><blockquote><strong><em>WHAT THIS MEANS PRACTICALLY</em></strong></blockquote><blockquote>Set money aside early! I applied in the 2022 cycle. I began saving for the Application Process three years earlier. I would suggest saving at least two lakhs per application cycle (at exchange rates of Rs. 82 per dollar). Saving more than that is a bonus — it gives you the space to apply to more universities.</blockquote><blockquote>This is also one of the reasons I recommend working for 2–3 years after your graduation before you apply for a PhD. There are other, better reasons to work before your PhD (which I’ll talk about another time), but a simple concrete one is that a regular salary helps bring money in for the application process.</blockquote><blockquote>A personal opinion — I don’t think you should borrow money for the application process. Either you should be capable of paying for the applications yourself, or someone should be willing to sponsor you at no cost (like a relative or a well-wisher). If you cannot afford the application process right now, think carefully about whether you want to apply. You can always apply a few years later, when you’re financially better off.</blockquote><p><strong><em>(3) Be prepared for a new “part-time job”:<br></em></strong>Most PhD applicants will be applying while they are simultaneously working at a job or completing their undergraduation/Master’s degree. In both cases, you’ll have to balance one important priority — your job/existing education — against another (your PhD application). This is of course, assuming that you don’t have other responsibilities.</p><p>I’ve applied for a PhD twice. In both attempts, the application process felt like a part-time job. In fact, before my 2022 attempt, I’d also begun an attempt in 2021, but had given up halfway without applying — I couldn’t balance the demands of my work with those of the application. At the end of the 2021 cycle, I spent several months discussing options with my supervisors to redesign my workloads. Based on their advice, I went part-time at the beginning of the 2022 cycle and this gave me more breathing room to work on my application.</p><p>Treating the Application Process like a part-time job may sound extreme, but it’s very much like a job if you look at it carefully. You have to cold-email lots of people. You have to prepare strategies and plans for the next few years. You have to market yourself (as a suitable candidate). You have to look up and pursue funders (scholarships). And like any other job, you have to brace for failures and carry on. All of this takes pretty much the same time and effort as a small job.</p><blockquote><strong><em>WHAT THIS MEANS PRACTICALLY</em></strong></blockquote><blockquote>The US/Canadian application cycle for most universities runs from September to December/January, about 4–5 months. During these 4–5 months, you should mentally assume that you’ll be juggling two jobs together.</blockquote><blockquote>Plan your tasks and your time accordingly. If you’ve had experience juggling two or more jobs before, rely on that experience to work out what schedule works best for you. If not, I would suggest the 4–2 routine (or 4–3 routine) — plan so that your current job/education will take an aggregate of 4 days of a week and your Application takes 2 days in a week (3 if needed, but not recommended).</blockquote><blockquote>Most importantly, plan for rest/relaxation and stick to it. If you don’t have experience balancing multiple jobs, the stress can be overwhelming and you can end up with burnout. Do NOT compromise on your rest or relaxation. Just plan for it beforehand.</blockquote><p><strong><em>(4) Make sure you’re REALLY sure :</em></strong></p><p>A PhD abroad is a big step. Not because of the syllabus or the rigour of the programme (those are important for different reasons), but because it is a significant commitment of your time, money, and resources. If you’re doing a British or European PhD, that’s at least 3 years. An American or Canadian PhD can go up to five or six years (maybe even more). That’s a fairly big chunk of your life. So before you apply, make sure you’re doing it for right reasons.</p><p>Personally, I feel you should NOT do a PhD for the following reasons:</p><p><strong><em>(a) Getting a “Dr.” before your name:</em></strong> You may laugh, but I’ve lost count of how many people I’ve met in India who chase a PhD for this. Don’t do it for this. No one cares about the Dr. before your name. Least of all, other PhDs.</p><p><strong><em>(b) Everyone else in your field is doing it: </em></strong>Pursuing a PhD because everyone in your field of work is pursuing a PhD is probably one of the worst (but most common) reasons to pursue a PhD.</p><p>In my opinion, any industry that <em>requires</em> you to have a PhD to do your job is asking a lot from you. Perhaps too much. If the only reason you’re doing a PhD is because you’re being peer-pressured into it, you might want to think about changing jobs or sectors.</p><p>Note: Not all peer pressure is bad. Some amount of peer-pressure is inevitable in the decision to pursue a PhD and it can sometimes encourage you to pursue one when you’re feeling hesitant. However, peer pressure should not be your <em>only </em>reason.</p><p>(c) <strong><em>You can’t think of anything else to do</em></strong>: Another terrible reason that’s unfortunately quite common. Don’t do a PhD just because you couldn’t think of any other options after you completed your Master’s degree. A PhD is not worth the time and investment in such cases. Put the same time and effort into finding new avenues of work and passion for yourself. That pays off better in the long run.</p><blockquote><strong><em>WHAT THIS MEANS PRACTICALLY</em></strong></blockquote><blockquote>Everyone has their own special reasons for applying to a PhD abroad. From my observations, the ones who do well are the ones who see the PhD as the next logical step of a much bigger, concrete set of goals in their lives.</blockquote><blockquote>And I mean <em>concrete</em>. The best PhD scholars I’ve met have very real-world concerns that they wish to tackle or solve. Pursuing a PhD is a way for them to find the space, the time, and the resources to think about how to approach their problems in ways they normally cannot in a regular job or industry. The best scholars complete their programmes with fairly clear plans of how they want to build on their research after their dissertation is complete.</blockquote><blockquote>My advice is to apply for a PhD only if you see it playing a clear, concrete role in helping you achieve bigger goals in life. Be clear about how doctoral training and research will help you with these bigger goals. But be aware that there is more than one way to achieve big goals or tackle big issues. A PhD is not mandatory for you to get there.</blockquote><p><strong>FINAL NOTE FOR PART 1</strong></p><p>A lot of what I said above may sound depressing, gloomy, or intimidating. And sure, there will be phases in the Application Process where everything feels that way.</p><p>But there’s also a lot of joy and pleasure in the process. A PhD application is one of those rare bureaucratic processes which allow you to turn the spotlight on yourself.</p><p>If you plan it well, you’ll get to really think about who you are and what the work you do actually means, for both yourself and the world around you. You’ll get to write about this, and present this to people who (hopefully) think and feel about the world in ways similar to you. One of the nicest feelings in the world is getting to hear a senior scholar tell you how much the faculty in their department enjoyed reading your statement and how it resonated with their own work.</p><p>However, the first step to getting to that point is to be aware of the work it involves. I’ve done my best to be honest and blunt about the workloads of a PhD application. By doing so, I hope I’ve given you some idea of what to expect, so that you can plan your own application accordingly. Keep in mind, that the difficulties may vary from candidate to candidate. An unusually brilliant student may find the Application Process a lot easier than what I’ve mentioned above. However, I do believe what I’ve written applies to many typical candidates, who are not outliers.</p><p>The very last thing I’ll say (from personal experience!) is to not get disheartened, disillusioned, or angry in the middle of the process. There are lots of problems with modern universities (and especially the application processes), anywhere in the world. But the middle of the process is not the time or place to dwell upon them. Make sure you work your way through the process, right up to the point you receive your decision letters. That’s the only way you can give it your best.</p><p>More practical steps in my next post.</p><p><em>A note about AI art</em>: I sometimes use AI art in my blogposts, strictly for content I put out for free, where I wouldn’t have hired an artist anyway. I use free AI platforms, or free options, and share prompts/sources so that anyone else can create AI art. Whenever I have time, I draw instead. I <em>do not</em> support the use of AI art when it displaces human artists.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b25e8f694f0f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Life and Career Update: Move to the United States]]></title>
            <link>https://amogharakali.medium.com/life-and-career-update-move-to-the-united-states-7f57b2aef88e?source=rss-17a040da3940------2</link>
            <guid isPermaLink="false">https://medium.com/p/7f57b2aef88e</guid>
            <category><![CDATA[public-policy]]></category>
            <category><![CDATA[doctorate]]></category>
            <category><![CDATA[phd]]></category>
            <category><![CDATA[urban-policy]]></category>
            <dc:creator><![CDATA[Amogh Arakali]]></dc:creator>
            <pubDate>Sat, 09 Sep 2023 18:06:15 GMT</pubDate>
            <atom:updated>2023-09-09T18:26:18.716Z</atom:updated>
            <content:encoded><![CDATA[<h4><strong>Commencing a PhD in Public and Urban Policy at the New School, New York.</strong></h4><p>If you’re following me on other platforms, you may have heard about this already, but in case you haven’t, I had a big life and career shift in August 2023. On August 21st, I commenced pursuing a PhD in Public and Urban Policy at the Milano School of Public Policy, one of the Schools of Public Engagement at the New School, New York, USA.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*m8GHGwdcIpNlYay4VMZLNA.jpeg" /><figcaption>The New School’s main University Center in New York. My classes are in other buildings though.</figcaption></figure><p>In some ways, this is a huge shift for me. In other ways, it feels like a natural next step in a research career that is now 10 years old (Note: although I began working in 2011, I didn’t shift to research until early 2013). While I’ve considered a PhD before, a part of me had leaned towards remaining in Bengaluru and building a more experimental research career. However, such experiments come with their costs — in particular, being taken seriously as a researcher. When I slowly began hitting growth barriers, I knew it was time to take the plunge.</p><p>However, I don’t want anyone to think I was forced into the plunge. While I’m frustrated about academia mandating such high levels of sacrifice, there is another, bigger, part of me that <em>loved </em>(and continues to love) the idea of a PhD. After 12 years of work, I believe deeply in the value of on-ground practice and “doing things”. Still, I’m also convinced that as long as practitioners don’t give themselves the time to engage, and reflect deeply upon their arena of work, their on-ground practice will be superficial. While there are many ways to engage and reflect, a well-structured PhD program remains a great space to do so, to improve your work and yourself. This is my primary reason for applying.</p><p>That said, a PhD isn’t just “<em>chalo kar teh hai”</em>. I’ve only had two weeks of classes so far, and I’m already working 15–16 hours a day, juggling three courses (each demanding about 15 hours of work a week), seven written assignments, a research assistant-ship, and the logistics of setting up a new life in New York. Financial costs remain a long-term concern (though I am stable for now). I am worried about sudden changes throwing all my work to waste. If you’re planning to take the plunge yourself, please don’t assume an easy-going life.</p><p>On the positive side though, the PhD structure seems exactly what I was looking for. The classes and assignments have been great, and the syllabus addresses exactly what I was looking for at this stage of my career. The work is also much more well-structured and organized than my earlier, professional life. So far, I’ve been able to plan my work weeks in advance, without worrying about sudden, unplanned disruptions. I’ve kept my weekends completely free (knock on wood), to switch off and do other things. And expensive as it is, New York is a <em>great</em> city to be in. Overall, this is a tough program, but so far, it’s been very rewarding.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CaCq5JVB79q9pGpcOz1OGg.jpeg" /><figcaption>Despite the expense, New York can be a great city.</figcaption></figure><p>I can already see that whatever happens, I will change due to this program, as well as due to being in a different country. My opinions and perspectives will certainly transform, as I absorb new knowledge and develop new perspectives on the world. Some of my career goals may shift, as new opportunities open up, and others close. My personality may change, or grow into new territories. I’m hoping these will be mostly good changes, with few negatives. I’m already wondering how to best hold on to my (extremely valuable) learnings from Bengaluru, Karnataka, India, and Asia, while remaining open to new insights here (more on this some other time).</p><p>Wish me luck! I’ll need it :)</p><p>— — —</p><p><strong><em>Post Script:</em></strong></p><p>PhD application season began again on September 1st.</p><p>In the United States and Canada, universities will be accepting applications for their PhD programs from September to December/January (deadlines will vary per university). Results will come in late Feb/March.</p><p>I’ll put up 2–3 blog posts in the coming weeks which may be useful for potential applicants from India, drawing on my own experience last year. These will be entirely dependent on when I’ll be able to write them, so they may take some time.</p><p>Meanwhile, if you’re looking to apply this year, I’d suggest starting your process now (if you haven’t already). You’ll need the full 3–4 months available, to prepare and send in a good application.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7f57b2aef88e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Numbers and Plans]]></title>
            <link>https://amogharakali.medium.com/numbers-and-plans-ea80c36394f4?source=rss-17a040da3940------2</link>
            <guid isPermaLink="false">https://medium.com/p/ea80c36394f4</guid>
            <category><![CDATA[china]]></category>
            <category><![CDATA[india]]></category>
            <category><![CDATA[statistics]]></category>
            <category><![CDATA[planning]]></category>
            <category><![CDATA[1950s]]></category>
            <dc:creator><![CDATA[Amogh Arakali]]></dc:creator>
            <pubDate>Sun, 05 Feb 2023 08:01:11 GMT</pubDate>
            <atom:updated>2023-02-05T08:01:11.795Z</atom:updated>
            <content:encoded><![CDATA[<h4>Reading Two Books, on Economic Planning and on Statistics in 1950s Asia (Long Post | #<a href="https://amogharakali.medium.com/list/frommyreadings-42795d91c322">FromMyLongReadings</a>, Issue 1, 2023)</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CZUYWTQf6WjQg-VwB5E5-w.png" /><figcaption>AI Image (MidJourney). Prompt*: “Indian Bureaucrats prepare a Five Year Plan in the 1950s”. Reality was far more complex than this. Also, I don’t know why there is a map of Africa at the back and why everyone has six fingers.</figcaption></figure><p>— — — — — — — <br><strong><em>#FromMyLongReadings is a new extension of my </em></strong><a href="https://amogharakali.medium.com/list/frommyreadings-42795d91c322"><strong><em>#FromMyReadings</em></strong></a><strong><em> blog series. Here, I’ll write longer posts (3500 words or more) about books and research papers I’ve been reading. Sometimes, I may also write shorter, more opinionated pieces to accompany these long posts.</em></strong></p><p><strong><em>Books for this post:<br></em>(1) </strong><a href="https://www.cambridge.org/core/books/planning-democracy/86CC2BAAF782ECFB0CAB433F94D5EF14"><em>Planning Democracy</em></a><em> </em>by Nikhil Menon <strong><br>(2) </strong><a href="https://press.princeton.edu/books/hardcover/9780691179476/making-it-count"><em>Making It Count</em></a><em> </em>by Arunabh Ghosh</p><p><strong>NOTE: Some illustrations in this post are AI-generated. These have been marked as such.</strong></p><p>— — — — — — — <br><strong>Introduction</strong></p><p>The end of the Second World War triggered a new chapter in history. In the Global North, the War’s after-effects pushed centres of power out from Europe, to the Soviet Union and the United States. In Asia and Africa, crumbling colonial empires gave way to dozens of new nations — independent, but confronting shaky futures. As the world reshaped itself, countries had to ask themselves how to move forward in uncertain times.</p><p>Most of us today roughly know what choices the world made — consolidating the United Nations; Reconstructing Europe; Developing the ‘Third World’; engaging in the Cold War; setting up the Non-Aligned Movement. Yet, ‘roughly’ knowing about these may not be sufficient.</p><p>The 2020–21 Pandemic may mark a turning point in history similar to what the world witnessed after 1945 (and also after 1991, but that’s for another day). At such turning points, the question of “How do we move ahead?” becomes crucial. New ideas will be proposed, and old ideas recycled. The future will seem exciting, uncertain, and unknown. At such times, it’s useful to look back (deeply) at other turning points in history, and learn from them. In this long post, I’ll use two books to look back at the 1950s. I’ll follow this up with a second, shorter post later on, examining what lessons the 1950s hold for us.</p><p>For this post, I was lucky enough to come across two books at the same time, which dive deep into 1950s India and China. The first book is <em>Planning Democracy</em> by historian <a href="https://history.nd.edu/people/nikhil-menon/">Nikhil Menon of Notre-Dame University</a>, published in 2022. The other is <em>Making It Count</em> by historian <a href="https://scholar.harvard.edu/arunabh.ghosh">Arunabh Ghosh of Harvard University,</a> published in 2020. While <em>Planning Democracy </em>focuses on independent India’s first attempts at designing Five-Year Plans, <em>Making It Count </em>focuses on early attempts in the People’s Republic of China to build national statistical systems.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*IetseeS2unhqSwpcV-221Q.jpeg" /><figcaption>The cover of <em>Making It Count </em>by Arunabh Ghosh</figcaption></figure><p>The two books aren’t perfectly comparable. One focuses on planning systems, the other examines statistical ones. Economic Planning and Statistics are linked to each other (they were particularly close in the 1950s), but they’re not identical. Still, both books are histories of national-scale attempts, by Asian countries of similar size and population, to build new technical systems and processes at a time of great change. There are enough overlaps between them for us to examine them together.</p><p>Both India and China underwent massive social and political transformations in the late 1940s (independence in India, revolution in China), which created new moments to change their future paths. In both countries, new laws had to be written, new programmes designed, new organisations built, and new systems put in place. If these changes were to be effective, decisions had to be consolidated, centralised, and tracked. Therefore, numbers and plans began to occupy the centre of the stage. Disciplines like Economic Planning and Statistics started attracting attention in political and bureaucratic circles.</p><p>It wasn’t long before technical experts in these subjects would become key figures of the 1950s.</p><p><strong>The Rise of Mahalanobis</strong></p><p>One person who dominates narratives in both books is Prasanta Chandra (P C) Mahalanobis. A physicist by training, Mahalanobis began forays into Statistics while studying at Cambridge University, something he continued at Presidency College in Calcutta. In a stroke of serendipity, Mahalanobis began his shift from Physics to Statistics precisely when the need for statisticians was rising among government planners. As Nikhil Menon explains in <em>Planning Democracy</em>:</p><blockquote>In the mid-twentieth century, several nation states witnessed a spurt in the production of statistical data.</blockquote><blockquote>[…]</blockquote><blockquote>Planners saw the economy as being essentially calculable.</blockquote><blockquote>[…]</blockquote><blockquote>Centralised planning required data about the economy, and statistics came to be viewed as the discipline that could deliver it.</blockquote><blockquote>- Planning Democracy, <em>Chapter 1, Page 6.</em></blockquote><p>Riding the tide, Mahalanobis set up the Indian Statistical Institute (ISI) in 1931 and launched a Statistics journal, <em>Sankhya,</em> in 1933. Over the next twenty years, both he and the ISI would build a reputation for themselves as the ultimate Indian authorities on Statistics (and later, Economic Planning).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/410/1*yFJqn4XJMm1ufZNW9ZOAuQ.png" /><figcaption>Cover of <em>Planning Democracy </em>by Nikhil Menon. P C Mahalanobis is sitting on the right.</figcaption></figure><p>The cornerstone of Mahalanobis’ and ISI’s reputation was the<em> Random Sample Survey</em>, a relatively new type of statistical study at the time. As the name suggests, statisticians choose a random sample of the full population under study, measure the sample, and extrapolate findings from the sample to the full population. The ISI pioneered the use of random sample surveys in Indian statistical practice, pushing them to become a critical tool of studying society and economy in the mid-20th Century. Random sample surveys would eventually become the basis for independent India’s National Sample Survey rounds, which commenced in 1950.</p><p>Additionally, Mahalanobis spent an enormous amount of effort burnishing his own reputation in global circles, travelling from conference to conference, meeting expert after expert in both the west and the Soviet Union. Menon claims there was a larger logic to his travels. Mahalanobis was an established statistician, but not a recognised economist or planner. By building connections with recognised economists and planners abroad, he could skirt around any questions about his credibility at home.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uF1eM6LT5UWvEUYdSPm8_Q.png" /><figcaption>AI Image (MidJourney). Prompt: “Statistician at Calcutta Airport in the 1950s, carrying a briefcase, travelling to an international conference.” I don’t know if the ISI’s statisticians wore hats like that — the AI seemed pretty insistent on retaining the hat, no matter how iterations I tried. Also, is that an aeroplane or metrorail?</figcaption></figure><p>It’s here that we start seeing intersections with Arunabh Ghosh’s book, <em>Making It Count</em>. As Mahalanobis became a well-known international figure, the ISI campus at Baranagore (near Calcutta) became a frequent stop for intellectual superstars and international dignitaries. The statistician Ronald Fisher; the economists Ragnar Frisch, Simon Kuznets, and Jan Tinbergen; the scientist J B S Haldane; American ambassador J K Galbraith; and Indonesian Vice President Mohammad Hatta had all visited the ISI in the 1950s.</p><p>In 1956, the Chinese Premier Zhou Enlai paid a visit. Zhou’s interest in the ISI’s work threw the “day’s itinerary out of sync”, as he spent his time getting involved in intensive discussions with the ISI staff and Mahalanobis. After extensive questioning on Indian statistical methods and practices, Zhou told Mahalanobis that a group of Chinese statisticians would visit ISI after his visit. Two days later, a team did visit the ISI and stayed for a month.</p><p><strong>The Story from China</strong></p><p>Chinese interest in the ISI’s work had quite a history behind it. Ghosh notes that although statistical work in China was already covering a wide range of activities by the 1930s, the work was fragmentary and limited. Though there were efforts to co-ordinate statistical work into a centralised system, these efforts were disrupted by the <a href="https://medium.com/m/callback/email?token=531473156ac2&amp;operation=login&amp;state=medium&amp;source=email-17a040da3940-1675411681175-auth.login------0-------------------e5fd77f4_f662_4c5a_893d_0f839bb3204d">1949 Revolution</a>, and rise of the Communist Party of China (CPC or CCP).</p><p>According to Ghosh, the CPC began expanding its own systems of statistics and data collection via its Northeast Statistics Bureau (NSB) in 1950. Both inspired and advised by the Soviet Union, the statistical work pioneered by the NSB and its director, Wang Sihui, strove for three basic features:</p><blockquote>(1) For statistical work to be <strong><em>extensive</em></strong>, by including data from every major component of China’s economy, culture, education, and health care.</blockquote><blockquote>(2) For statistical work to be <strong><em>complete</em></strong>, so that whatever data *could be* captured *was* captured, and standardised into a central system.</blockquote><blockquote>(3) For statistical work to be <strong><em>objective</em></strong>, so that “unlike in capitalist systems”, statistics was not distorted or censored by profit motives and would not hide weaknesses.</blockquote><p>In order to fulfill these criteria, Chinese statistical systems adopted two means of collecting data. The first was <em>Exhaustive Enumeration. </em>Simply put, this meant collecting data about everything deemed important — as broadly, deeply, and as frequently resources would allow.</p><p>The second way of collecting data (especially in agriculture) was via <em>Typical Sample Surveys</em>. Unlike the sample surveys pioneered by the ISI in India, samples here were selected by specified criteria instead of random selection. In addition to these two methods, there were several other forms of statistical work, from calculating price indices to correcting old records.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/531/1*BxuZLOFJ0asyniYS9oiBIg.png" /><figcaption>AI Image (MidJourney). Prompt*: “Statisticians at work in the People’s Republic of China in the 1950s”. I have no clue what these two gentlemen are supposed to be doing.</figcaption></figure><p>These lofty goals came with a toll. The Everything-Everywhere-All-At-Once approach adopted by the Chinese statistical systems began to generate far more data than could be processed by the people at the top. Reports started coming in of regional offices flooded with excess printed data which had no place to go. Additionally, discrepancies between local, regional, and national data began to increase, as tight deadlines pushed statisticians at higher levels of government to rely more on estimates than enumeration. A unified national statistical network was set up in the 1950s, but problems persisted.</p><p>Furthermore, Chinese statisticians were beginning to experience an ideological crisis. Although the Soviet Union had advised and supported the development of statistical systems in the early days of the People’s Republic, the Soviets were being viewed as unworthy of emulation by the mid-1950s. In 1956, Zhou Enlai commented that Chinese study of the Soviet Union had been conducted with “undue haste, arbitrary learning, and mechanical application”. By this time, China was opening itself to new perspectives from other countries, including India.</p><p><strong>A Brief Time Together</strong></p><p>In some ways, the statisticians of India and China were worlds apart when they met in 1956.</p><p>Following the Soviets’ example, the People’s Republic of China had wrung statistics through an ideological mangle. A culture of criticising the ‘bourgeois statistics’ of the West had entrenched a belief that statistics was a social science <a href="https://www.journals.uchicago.edu/doi/10.1086/699237#_i3">with no universal rules</a>. Ghosh dwells quite a bit on China’s suspicion of statistics which was rooted in formal mathematics, since it was seen as conflating social and natural worlds together. Concepts like <a href="https://en.wikipedia.org/wiki/Central_tendency">Measures of Central Tendency</a> (Mean, Median, Mode) or <a href="https://www.bea.gov/help/glossary/fisher-index">Fisher’s Price Index</a> were criticised as distorting social reality. Overall, the People’s Republic of China had spent years distancing itself from the statistical practices of capitalist countries.</p><p>By contrast, Mahalanobis and the ISI had spent decades engaging with the west (and elsewhere) to legitimise their work on random sample surveys. When appointed as a chair for a new UN Sub-Commission on Statistical Sampling, Mahalanobis advocated for spreading sampling methods worldwide. Menon mentions that India’s National Sample Surveys became the norm for UN Household Surveys. Furthermore, Indian statisticians were less suspicious of what the Chinese called ‘bourgeois statistics’. For instance, India used random sample surveys to pioneer the concept of the Poverty Line in 1962, an idea that might have been criticised in China.</p><p>However, the changed attitude to the Soviet Union by 1956, and a new openness in China’s foreign policy after the <a href="https://en.wikipedia.org/wiki/Bandung_Conference">1955 Bandung Conference </a>meant that the Chinese showed tremendous interest in Indian statistical methods. In both Ghosh’s and Menon’s books, the initial meetings between the Indians and the Chinese were quite positive. Prior to Zhou Enlai’s visit to ISI, an Indian team led by <a href="https://en.wikipedia.org/wiki/Pitambar_Pant">Pitambar Pant</a> has visited China in July 1956. Ghosh describes a meeting with Zhou Enlai during this visit, where Zhou praised Indian statistical measurements of agriculture.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/680/0*r-my5BySjx3LUuEe" /><figcaption>Zhou Enlai on his 1956 visit to India. Here, he’s seen visiting Mamallapuram in Tamil Nadu. (<a href="https://www.deccanherald.com/national/south/when-chinese-premier-zhou-enlai-visited-mamallapuram-767210.html">Source</a>)</figcaption></figure><p>After the Chinese team visited ISI, Mahalanobis and his colleagues were also invited to lecture in China in 1957. Given China’s earlier suspicions of mathematical connections to statistics, it’s interesting to note that during this visit, Mahalanobis gave a lecture on <em>The Theory of Mathematical Probability in Sample Surveys</em>, while his colleague D B Lahiri delivered four lectures on <em>Applications of Mathematical Probability in Sample Surveys</em>. Even if Chinese suspicion of mathematical statistics hadn’t vanished, they were now willing to listen to alternative opinions.</p><p>This optimistic state of events wasn’t to last. Initially, there were some good signs. After Mahalanobis and his team returned to India, Ghosh reports how Wang Sihua of the NSB publicly advocated for adopting Indian random sampling techniques in Chinese statistical work. However, his advocacy was guarded, suggesting that random samples be used only when original records were missing or where numbers to be enumerated were too high. In other cases, Wang suggested that “a socialist country” like the People’s Republic of China could not abandon complete enumeration or Soviet methods entirely.</p><p>The larger political climate in China was much more disruptive. In June 1957, Mao Zedong and Deng Xiaoping commenced <a href="https://en.wikipedia.org/wiki/Anti-Rightist_Campaign">the Anti-Rightist campaign</a>, where a purge of CPC members perceived as not being left-wing enough was conducted. This was followed by Mao’s Great Leap Forward in 1958, where academics and intellectuals were evicted from cities and sent to work in agricultural communes. In such an environment, Ghosh notes that intellectual attempts to be open-minded to foreign ideas was to invite disaster.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*FMBx4M5bBMHxO4C_.jpg" /><figcaption>China’s 1957 National Day Parade at Tienanmen Square. The slogan “carry out the anti-rightist struggle to the end” is seen on display (Source: <a href="https://en.wikipedia.org/wiki/Anti-Rightist_Campaign#/media/File:Anti-Rightist_Movement.jpg">Wikimedia Commons</a>)</figcaption></figure><p>Both the existing Chinese statistical methods, (Ghosh notes) as well as the random sampling techniques from India, were sidelined during the Great Leap Forward. Statistical methods of the NSB had always sat uneasily with Mao’s own preferences of relying upon in-person experience, and on using singular examples as representative of larger trends. During the Great Leap Forward, larger surveys, sampling, and comprehensive data systems were ignored in favour of more personalised methods.</p><p>In addition to this, diplomatic relations between India and China had begun to fail. Negotiations on fixing the disputed Sino-Indian border broke down, and the 1959 exodus of the Dalai Lama to India effectively ended most academic and cultural exchanges between India and China. A few years later, in 1962, the two countries would go to war over the unresolved border. The statistical exchange between the countries had had a promising start, but it was cut short too soon to be of lasting value.</p><p><strong>Statistics in Five Year Plans</strong></p><p>India-China comparisons have been going on for years. They are the only countries in the world with billion-plus populations, they are both Asian, both are among the world’s largest countries by landmass, both have high economic growth rates, and both are expected to become geopolitical superpowers this century. There are significant differences of course, but there is still enough to warrant comparisons with each other. Studying them can often reveal interesting parallels.</p><p>I noted two such parallels when reading <em>Planning Democracy </em>and <em>Making It Count</em> together. The first was in the use of Statistics to build Five Year Plans. Both India and China carried out Five Year Planning exercises in the 1950s, both for the first time. In both cases Statistics was considered subservient to the larger goal of the Five Year Planning process.</p><p>It’s important for us to remember how ubiquitous the ideas of Economic Planning and National Development were in the 1950s. At several points in his book, Menon mentions how the notion of Planning had permeated the national consciousness in India. “The Five Year Plans…” he observes. “…represented an alternate national calendar.”</p><blockquote>“A satirical piece in <em>Shankar’s Weekly</em> noted that visitors to the country may notice that ‘India is very plan-minded’.”</blockquote><blockquote>[…]</blockquote><blockquote>“‘You cannot get away from this planning business…whatever you talk about somehow they [Indians] lead you to the question of planning as though everything depended on planning, everywhere.’”</blockquote><blockquote>- Planning Democracy<em>, Chapter 4, Page 114.</em></blockquote><p>In an atmosphere enveloped in Planning fervour, technical disciplines and scientific work had to be adopted to serve the larger national good. <a href="https://en.wikipedia.org/wiki/Vikram_Sarabhai">Vikram Sarabhai</a> once defended India’s space programme by commenting that “…we [India] must be second to none in the application of advanced technologies to the real problems of man and society”. Even to Mahalanobis, the relationship between Statistics and Economic Planning was clear. Menon notes how Mahalanobis once commented “Poverty is the most basic problem of the country, and that Statistics must help solving this problem.”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/531/1*X4HNRILQzr8atSxOUyhVwg.png" /><figcaption>AI Image (MidJourney). Prompt: “Statisticians and Economic Planners at a conference in Delhi in the 1950s.” Please remember that this is an AI-generated illustration created for fun, not to be taken as reality.</figcaption></figure><p>This relationship between Statistics and Economic Planning in India had also been formalised. In 1950, the Government of India set up the Central Statistical Organisation (CSO) in tandem with the establishment of the country’s Planning Commission. While the Planning Commission was in charge of the Five Year Planning Process, the CSO would organise the statistical apparatus of the Government to serve the Planning process. Mahalanobis was regularly invited to Planning Commission meetings and the CSO shared physical space with the Commission at Yojana Bhavan in Delhi. The ISI’s workhorse, the Random Sample Survey served as the statistical backbone of the Planning process, with the first National Sample Survey in 1950–51 covering 1833 villages in 15 languages (translated back to English).</p><p>A similar attitude to statistics and planning prevailed in China. The Five Year Planning process in China (at the time) was co-ordinated by its State Planning Commission or SPC. After preparing and circulating a draft plan, the SPC would receive smaller draft plans from lower-level organisations in the government hierarchy which would then be used to revise the national plan.</p><p>At the lowest levels, Ghosh notes that Planning and Statistics were combined into a single office. However, at the national level, the State Planning Commission would partner with the State Statistical Board or SSB, in a manner not too different from how India’s CSO worked with its Planning Commission. The SSB and SPC occupied the same building, just like the CSO and the Planning Commission in India.</p><p>In an echo of Mahalanobis, statistical work in China was seen as serving the national planning process (Ghosh observes this could have stemmed from Mao Zedong’s emphasis to “serve the people”). Statistical numbers were the basis for preparing plans, inspecting them, deciding on policies and checking on implementation. In 1954, the SPC released a six-page document which exhorted for greater co-operation and understanding between the Planning and Statistics offices in order to solve pending issues of co-ordination.</p><p><strong>The Role of ‘Ideal People’</strong></p><p>The second parallel which I noted in both books was how both India and China relied upon idealistic depictions of ordinary people to make their respective systems work. It’s important to note that the two countries had (and continue to have) very different political systems. Yet, both regimes had to ensure that their plans and systems were considered legitimate by the people at the grassroots. Both regimes did this by creating imagery of ordinary people as integral parts of the larger planning or statistical processes. By creating such imagery, both nations sought popular support for their national initiatives.</p><p>Menon observes that in independent India, citizen rights were often seen as coming attached with duties, to be performed by the citizens for the nation. Very often, this would require ordinary citizens to become part of the planning process in some way or another. Menon argues that state campaigns in India to remind Indian citizens of their duties were crucial to to Plan success, since “unlike in the Soviet Union or China”, the Indian government could not force its citizens to comply.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/587/1*UHs6Ie2PLxOYHkzFSvbtMQ.png" /><figcaption>AI Image (MidJourney). Prompt: “Indian workers and engineers engaged in building a steel plant in a small Central Indian town in the 1950s”. Unlike in this image, I hope none of the buildings in <a href="https://en.wikipedia.org/wiki/Bokaro_Steel_City">Bokaro</a> or <a href="https://en.wikipedia.org/wiki/Rourkela_Steel_Plant">Rourkela</a> caught fire.</figcaption></figure><p>While it is true that the People’s Republic had far more stringent means of ensuring public participation, it does not mean that the Chinese government didn’t face its own challenges. Ghosh observes that sometimes the hardest people to convince were ground-level workers within the Chinese bureaucracy itself. In Chapter 6 of his book (wryly titled <em>To Ardently Love Our Statistical Work)</em>, Ghosh talks about how (unlike engineering or farming) statistics was not seen as “productive work” by ground-level statistical workers, since it did not produce tangible goods and dealt with abstract notions and concepts. There was a need to communicate the importance of statistical work and its role in nation-building.</p><p>Authorities in China undertook a slew of initiatives to train ground-level workers in both methods and the importance of statistical work. Apart from the many training institutes and courses however, a significant number of debates and discourse also took part among the pages of statistical journals, with (likely fictional) ground-level workers lamenting on the uselessness of statistics and experts providing responses.</p><p>From these discourses arose the romantic idea of the Ideal Statistical Worker. Ghosh writes that lamentations about the uselessness of statistical work got responses about “needs of the nation and the people”. Statistical workers were criticised for thinking only about their personal prospects and profits and told to focus on the bigger picture. On the other hand, ‘good’ statistical workers began to receive ‘model worker’ awards. In May 1956, nearly 300 people were honoured as model workers in a national conference. Furthermore, propaganda painted statistical reports as tangible, desirable outputs, thereby addressing the claim that Statistics was abstract, useless work.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/531/1*kCRluIfO5TELnOhTgHWqEw.png" /><figcaption>AI Image (MidJourney): Prompt: “The ideal statistical worker in the People’s Republic of China in the 1950s”. I cannot read Mandarin, so I REALLY hope the AI hasn’t generated any that is rude or offensive (please let me know if it is). The limits of using AI for illustrations are becoming rather apparent.</figcaption></figure><p>At the same time, in India, massive public outreach campaigns were being designed to raise awareness of and participation in the Five Year Plans. Much of the second half of Menon’s book focuses on Indian government attempts to increase public participation in the Five Year Plans. He reports on how Nehru commented that the Planning Commission’s task also involved “…how to get the villager…to realise that something is not only being done for him, but he’s doing it and is part of the machinery doing it…” The Indian government created mass campaigns at the 1954 <a href="https://en.wikipedia.org/wiki/Kumbh_Mela">Kumbh Mela</a>, set up a Five Year Plan Publicity Organisation, distributed pamphlets and posters about “Help the Plan, Help Yourself”, held exhibitions for urban citizens, and released a documentary.</p><p>The parallels between Indian citizens participating in the Five Year Plans and in Chinese statistical workers participating in the national statistical work are therefore quite striking. In both circumstances, an ideal worker or a citizen was someone typically from the grassroots or lower levels who may not have truly understood the scope or the scale of the project they were ostensibly part of, but who performed their duties for the national good nonetheless. In both circumstances, it became the duty of senior bureaucrats to engage with the grassroots, promote images of the grassroots worker as part of a larger national agenda, and recruit them to larger causes.</p><p><strong>Conclusion</strong></p><p>Reading Menon and Ghosh together brings out how crucial the 1950s were as a decade of change for Asia as a whole. There were other critical moments as well — the 1955 Afro-Asian Conference at Bandung for instance — but the fact that two of Asia’s largest countries embarked upon drastic political, bureaucratic, and economic restructuring at the same time effectively make the 1950s a turning point for Asian history. Much of what we see in India and China today took root during that decade.</p><p>Furthermore, despite the deep differences in ideology and political structure, despite the trigger points for change being so vastly different in the two countries (independence from a colonial power versus ascendancy via armed revolution), it’s remarkable to see broad similarities in the challenges the two countries faced, and in the ways they chose to respond. They even interacted with each other briefly about these, before deeper issues destroyed relations between the two.</p><p>I’ll need a separate post to ruminate on the implications of the 1950s for our present times, but some brief points may be useful here. Firstly, it’s useful to remind ourselves that neither India nor China are what they used to be in the 1950s. In India, the Planning Commission and the Five Year Planning process were abandoned in 2015 for being too archaic, while the country’s national statistical architecture is currently under reform. In China, sweeping changes under Deng Xiaoping and more recently, under the Xi Jinping regime, has made it a very different nation from Mao’s and Zhou’s time. For both countries, private sector growth, corporate-led capitalism, and the emergence of a consumption class has come to signify popular markers of growth.</p><p>That being said, there are some echoes between what Menon and Ghosh write about the 1950s and what we see today. We may no longer harp about enumerations or random sample surveys to the same extent as back then, but the need for large-scale statistics continues to remain high. Debates now focus on Big Data, Data Science, and Artificial Intelligence to answer deep questions about our society in similar ways. We may no longer resort to Five Year Plans, but grand visions built on rising economic prosperity continue to dominate discourse, with governments in both India and China finding new ways to tap into global economic trends.</p><p>In some ways, it does seem like these two countries have come far from the days covered by Menon and Ghosh. In other ways, we seem to still be right where we began. More on this in a later post.</p><p>— — — — — — — — — — — —</p><p><em>*A note on the AI-generated images: </em><a href="https://intellectual-property-helpdesk.ec.europa.eu/news-events/news/ai-art-and-copyright-protection-india-2022-11-04_en"><em>The copyright status on AI-generated images in India is still unclear.</em></a><em> In the absence of a clear legal judgement, I’ll be treating AI software like DALL-E and MidJourney as tools, and claiming any art work generated through my prompts to them as my own. (I’ll update this note if the legal status becomes clearer).</em></p><p><em>However, there should be no contest that </em><strong><em>the prompts</em></strong><em> I give AI software are my own. I request that if you make AI art of your own by using the same prompts I’ve used in this post — </em><strong><em>can you please acknowledge me as the original author of those prompts?</em></strong><em> Just a note in your final publication will do. I’m not going to charge you or anything, I‘d just appreciate the acknowledgement! Thanks.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ea80c36394f4" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How Do AI Represent the Urban?]]></title>
            <link>https://amogharakali.medium.com/how-do-ai-represent-the-urban-d439b4c571f5?source=rss-17a040da3940------2</link>
            <guid isPermaLink="false">https://medium.com/p/d439b4c571f5</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[cities]]></category>
            <category><![CDATA[urbanization]]></category>
            <category><![CDATA[urban]]></category>
            <dc:creator><![CDATA[Amogh Arakali]]></dc:creator>
            <pubDate>Sat, 04 Dec 2021 11:12:21 GMT</pubDate>
            <atom:updated>2021-12-04T13:09:44.299Z</atom:updated>
            <content:encoded><![CDATA[<h4>Using an Ongoing Fad to Ask About Artificial Intelligences and Urbanisation</h4><p>I’m on a break from regular blogging until my teaching and research-writing responsibilities for the Financial Year end in March. However, I’m briefly dipping back to talk about something interesting.</p><p>This week, my social media feeds were filled with blurry abstract paintings. I follow many authors and artists on my personal handles. Several of them had gotten excited about this new AI app (called <a href="https://www.wombo.art/">Wombo Dream</a>) that generates art for you when prompted. For instance, science fiction author <a href="https://twitter.com/jdiannedotson/status/1466819892028477446">J Dianne Dotson got the app to design potential book covers for her</a>. Here’s my own attempt to create an image for <a href="https://amogharakali.scrollstack.com/post/4795/-1-The-River-in-the-Sky-and-the-Banyan-Tree-Part-1-">a short story I’d written earlier this year</a>:</p><figure><img alt="AI-generated image for the prompt “The River in the Sky and the Banyan Tree”" src="https://cdn-images-1.medium.com/max/1024/0*lgiWvKlcr8LyFXVq" /><figcaption><a href="https://amogharakali.scrollstack.com/post/4795/-1-The-River-in-the-Sky-and-the-Banyan-Tree-Part-1-">The River in the Sky and the Banyan Tree</a> (Wombo Dream Illustration). By: Amogh Arakali, 2021</figcaption></figure><h4>Playing Around with Urbanization</h4><p>At this moment, I decided to try something related to my work. Mind you, I wasn’t conducting any deep study. All I did was open the app and enter some prompts. The prompts were fairly straightforward:</p><ol><li>“Indian Urbanization”</li><li>“Chinese Urbanization”</li><li>“South African Urbanization”</li><li>“Nigerian Urbanization”</li><li>“Egyptian Urbanization”</li><li>“Indonesian Urbanization”</li><li>“Japanese Urbanization”</li><li>“European Urbanization”</li><li>“American Urbanization”</li></ol><p>Here are my results:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Dv51gzR_AVpETOW0Gsemrg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OKY7tmByNs9r4m6iJ9jbQg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3UbR_xR9XPOKBYnFu9bjZQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*P6ximI2-R8mej7ERMrUpsQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eiZccTdx6wctJkCG-s4sJw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BpZaR-47s3jDaiLSyjjRyQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aCJTqlD2L_grO7BBTGCknw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wy9h_jVqA8wyocsD_H-EtA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ohENrAZXkclJ_e_ZmId3pw.jpeg" /></figure><p>These are not actual places or locations in the real world. These are not even real-life objects. Instead, they are unique combinations of patterns which have been strung together by a programme. It’s true that these patterns are derived from real images, but it’s a bit hard to argue that they are ‘real’. It’s like human beings trying to redraw their childhood home on paper. Something will be off-kilter, something will not quite fit with what actually existed in real life. And yet…and yet…the patterns will show us <em>something</em> we can recognise.</p><p>Take the South African one for instance. You’ll never find an exact image like that from a South African city. But compare the AI-generated dreamscape with these pictures from South Africa below. There’s <em>something </em>that’s familiar across these pictures, but I hesitate to label it:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3UbR_xR9XPOKBYnFu9bjZQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/266/1*-Y8_5oxg3camLGmQ-KZDqw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*eEH9kxOsyfzgBmCH91GV2g.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1vNpciCim8dIq8Hsd-prWw.jpeg" /><figcaption>Clockwise from top left: The Wombo AI-generated image (Amogh Arakali using Wombo.art); a satellite image of Southern Africa (Wikimedia Commons); A photo from Cape Town (Wikimedia Commons); The South African Flag (Wikimedia Commons)</figcaption></figure><h4><strong>AI Biases</strong></h4><p>It’s important to remember that these images aren’t created from scratch. They’re built from “training sets” of images that human researchers feed into the AI to help it learn and recognise patterns. If you’re not familiar with how such AI apps work, t<a href="https://www.theguardian.com/technology/2015/jun/18/google-image-recognition-neural-network-androids-dream-electric-sheep">his old article from 2015 does a pretty good job of explaining this</a>. I suspect that while the process has become more sophisticated over the years, the basic principle of recursively feeding images back into neural nets until the AI “gets it” hasn’t changed.</p><p>Therefore, human biases do exist in the patterns chosen and images generated, which turns these images into AI interpretations of human biases. If too many people are biased towards photographing and uploading brightly-coloured buildings in South African cities onto the training sets, it shouldn’t be too surprising when AI replicate the same pattern.</p><p>But the problem is that it’s really hard to prove that <em>this</em> bias is what drove the AI’s choice to have several colours in the final image. After all, colours are associated with South Africa in more ways than simply the shades of its buildings. Did the AI include multiple colours because of the houses in Cape Town, the multiple colours of the South African flag, or because of the long associations between colours and politics in South African history (it is The Rainbow Nation after all)? Or was it something else altogether? It’s simply too difficult to tell, as an end-user.</p><h4><strong>Human Biases</strong></h4><p>Urban scholars have long wrestled with the question of ‘representation’ — how certain places, communities, people, and cities get represented in urban imagery and media. There are old arguments which point out that that cities like New York, London, Paris, Tokyo (and in more recent years, Singapore and Dubai) are over-represented in our imaginations of what the “ideal urban” actually is, and urban ‘solutions’ from these places are assumed to hold for most other cities.</p><p>On the other hand, there isn’t enough attention on what’s happening in these “elsewheres”. Some of the largest and fastest growing cities in the world — Bengaluru, Shenzhen, Lagos, Jakarta, Delhi, Cairo — barely feature in global urban discourses. Knowledge about other Asian, African, and South American cities (Indore in India, Tamanrasset in Algeria, Belo Horizonte in Brazil) is pretty much non-existent.</p><p>I was curious to see if an AI app like Wombo Dream could add to this discourse in some unique way. After all, visual art is one of the most powerful methods of representation. Imagery from <em>Tron</em>, <em>Blade Runner</em>, <em>iRobot </em>and <em>Minority Report</em> defined how people in power visualised futuristic cities. Such imagery has influenced how cities like <a href="https://www.youtube.com/watch?v=RkNiDKR_hc8">Dubai represent themselves in the future</a>. Do AI add anything new to this discourse?</p><h4>Diving a Bit Deeper</h4><p>I decided to try one more picture before attempting anything else. I simply typed “Urbanization” to see what I would get:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pkkKytzfHTvQKF3PaRo5cw.jpeg" /><figcaption>Hmm…</figcaption></figure><p>This was a fascinating result. I got impressions of an urban fabric that feels very familiar (having pored over these in satellite image after satellite image):</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ms1i0Ij5u5Ne3xAY7AM-VQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*u-z9Sc0CxwNEf-MC_FUCOA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*obsB7vMO5928v51MqyJgLA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xLhHvctTDduK_YMVc5LI2w.png" /><figcaption>Clockwise from Top Left: Bahnaya (Egypt); Cox’s Bazaar (Bangladesh); Siwan (Haryana, India); Delhi (NCT, India)</figcaption></figure><p>Given technology biases to First World Countries, I had expected a landscape very similar to the “American Urbanization” or “European Urbanization” pictures above (or at the very least, a “Japanese Urbanization”). In contrast, here are some satellite images of American and British urban fabrics I’d expected in the image above:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bYmmQcqjms9Gyf2VcLBLBg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WphxFu8GQenb46tGIMDPuQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3M8Qv5wRo2W6yhuI9kI8Sw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EYoa_g1N5N3Q3tfu5cTnuQ.png" /><figcaption>Clockwise from Top Left: Merrick (Long Island, USA); Newark (New Jersey, USA); Wandsworth (Outer London, UK); San Francisco (California, USA).</figcaption></figure><p>Instead, I was getting something a lot closer to home here in India. I tried ‘Urbanization’ two more times to see if I would get a different result. I then tried “City”, “Town” and “Urban”. The results are included below:</p><p><strong>“Urbanization”:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pkkKytzfHTvQKF3PaRo5cw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NPQ2W_6AEGl9BC-s_kX_MA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ApkHY0HyERHK7QyuqX3kKA.jpeg" /><figcaption>Images by Amogh Arakali (using Womble Dream)</figcaption></figure><p><strong>“City”:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4LwnrBF7JXfuyBeBr-6aXw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-oyHTT19t4hJpMVTd8L4mA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KQ0MEZu54ei4w2ugm6vUVQ.jpeg" /><figcaption>Images by Amogh Arakali (using Womble Dream)</figcaption></figure><p><strong>“Town”:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TyRiRwwMggqlU_wpaA4Ieg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1tIFWZ8TFV8JWtSKD6_Nfg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5BgHkMR7e7XBvh4EGCKW3w.jpeg" /><figcaption>Images by Amogh Arakali (using Womble Dream)</figcaption></figure><p><strong>“Urban”:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-C7qbtXNn8lgFXWiuuw64g.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n9UCHn9h_mbYj0xgcbTWWw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*16wCK_2lsMrtoRRFUWvBKA.jpeg" /><figcaption>Images by Amogh Arakali (using Womble Dream)</figcaption></figure><p>This just got more and more fascinating. Briefly, here are some of my observations:</p><ul><li>There is a fair difference between how the AI interprets “City”, “Town” and “Urban”. While there are overlaps (see Town 3 and Urban 1 for instance), there is enough of a difference to assume that the AI treats them as separate categories.</li><li>“Urbanization” and “Urban” seem to be more strongly associated with dense, packed settlements, often (but not always) viewed from above. It’s likely that some of the training set images for these two categories came from satellite photos, most probably of Asian or African cities.</li><li>The AI seems to interpret “City” very distinctly, associating it with tall towers viewed from a distance, with lots of bright lights. The training set may have included multiple pictures taken at night, of large metropolises, most probably from the US or east Asia.</li><li>The “Town” category is a bit confusing. Two of the pictures are really similar, suggesting training sets of images from small European and American towns (the church steeple, a very European characteristic, features in both images). <br>However, the third threw me off completely, looking much more like an “Urban” image than a “Town” one. I’ll test this out more when I get the time, but it suggests to me that “Town” might be a fairly broad category, with multiple interpretations.</li><li>Not too surprisingly, the “Urban” category is closely linked to “Urbanization”. There are probably significant overlaps in the images they refer to, again implying links to dense settlements, most probably (but not confirmed) from Asia or Africa</li></ul><p>I should note that these aren’t very strong conclusions, because the sample size I’ve used is far too small. Ideally, I should be generating anywhere between 100–500 images for each of these categories (at least) and analysing commonalities across them. I shouldn’t also be referring to only one AI app, but multiple AI systems. That being said, these are some useful, quick insights which tell us where to go.</p><h4><strong>The Question to Ask</strong></h4><p>Again, I should reiterate that the AI is creating all its new images using pictures and images from human beings. The question I’d love to ask (but I’m not sure how to answer) is this — <strong><em>are the AI’s choices of images biased only by the researchers who determined the training sets, or are they also biased by how general populations perceive these categories?</em></strong></p><p>For instance, take the “city” category, with the tall towers and bright lights. How much of the AI’s choice to include towers was determined by:</p><p>(a) The researchers who decided that “City” <em>must</em> include tall towers?</p><p>(b) Peoples’ tendency to label a place with tall towers as “a City” when uploading an image to the web (which was then included in the training set)?</p><p>(c) A general attitude among people to associate tall towers with cities?</p><p>Will this differ from place to place? For instance, would people from South Asia or Central Africa be less inclined to identify towers with cities, given that many cities in those regions do not have skyscrapers? Or has the image of a city as having tall towers become universal enough for <em>most people</em> in the world, regardless of location, to associate Cities with towers?</p><h4>Going Ahead</h4><p>I am not sure how and to what extent the growing prominence of AI in the arts will impact industries like media, advertising, architecture, civil engineering and design. However, all these industries play important roles in shaping popular notions of urban, urbanisation, and cities.</p><p>If AI affects these industries to any significant extent at all, we should expect our imaginations of these concepts to be challenged and changed. Younger generations of people, with greater exposure to these new images are likely to encounter AI art around cities and urbanisation. If such a case does turn out to be true, then we’ll have to start engaging with both the good and bad of AI interpreting the spaces we inhabit. More importantly, we’ll have to watch out for future AI being manipulated by humans into specific interpretations.</p><p>It’s possible that AI art will die out quickly as a fad. After all, there have been technologies upon technologies in the past which have promised (or threatened) global dominance and failed. However, I’m not so sure. While I don’t think AI is ever going to replace human beings, it will change the way we build relationships with the environments around us, as well as with each other. If this is indeed the case, it’s best to start engaging with AI interpretations of the Urban as soon as possible.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-jgbRpowH49Ls4dekjuoxA.jpeg" /><figcaption>That’s a bit ominous.</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d439b4c571f5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Amogh Arakali’s #FromMyReadings, Issue 7, 2021]]></title>
            <link>https://amogharakali.medium.com/amogh-arakalis-frommyreadings-issue-7-2021-7400bee134c7?source=rss-17a040da3940------2</link>
            <guid isPermaLink="false">https://medium.com/p/7400bee134c7</guid>
            <category><![CDATA[indian-economy]]></category>
            <category><![CDATA[employment]]></category>
            <category><![CDATA[covid-19-lockdown]]></category>
            <category><![CDATA[unemployment]]></category>
            <category><![CDATA[india]]></category>
            <dc:creator><![CDATA[Amogh Arakali]]></dc:creator>
            <pubDate>Mon, 02 Aug 2021 17:40:30 GMT</pubDate>
            <atom:updated>2021-08-02T17:40:30.374Z</atom:updated>
            <content:encoded><![CDATA[<h3>#FromMyReadings, Issue 7, 2021</h3><h4>Urban Employment Data from and after the 2020 Lockdown</h4><p><em>#FromMyReadings is a series of short summaries of books, chapters, papers, articles, and media I come across in my regular work as a researcher.</em></p><p><em>I try and put up an issue of #FromMyReadings at least once a fortnight, going up to once a week when there’s time. Mid-August to March is teaching season, so my posts reduce/stop during this time.</em></p><h3>1. The Quarterly PLFS Bulletin: July-September 2020 (MOSPI, Government of India)</h3><figure><img alt="Bricklayer in India. Source: https://www.maxpixel.net/Loading-Bricks-Indian-Labour-Truck-Labourer-166919" src="https://cdn-images-1.medium.com/max/960/0*dXSQXL4e0zEnQ19N.jpg" /></figure><p>A few hours ago, the Ministry of Statistics and Programme Implementation (MOSPI) of the Government of India <a href="https://www.mospi.gov.in/documents/213904/301563//PLFS_Quarterly_Bulletin_Jul_Sep_2020m1627900970967.pdf/1b035713-c748-12fa-8de7-87da5b98c05c">released urban employment data for the 2nd Quarter of Financial Year (FY) 2020</a>, from July to September in a new Quarterly Bulletin. This is a particularly important quarter to study, given that it was immediately after the massive nation-wide lockdown which was imposed due to COVID-19 last year. My first impressions are recorded below.</p><p>Two things to note before anything else. First, the employment and labour force participation picture for India isn’t good <em>generally</em>. We know this already, but it’s also easy enough to see from the data in the Quarterly Bulletin. Labour Force Participation (LFP) for males above the age of 15 was below 75% even in 2019, well before the pandemic and the lockdown. For females, the LFP was far worse, hovering between 19-20%.</p><p>Second point: Large-scale national statistics like these need to be treated very conservatively. Unless there’s evidence for the opposite, it’s good to assume that the picture on the ground may be slightly different than what’s depicted on paper. Empirical data collected at large scales will also have flaws which are hard to detect/correct, there’s no getting around that. So what we have is a grim picture that may be worse than depicted.</p><p>It’s important to note that LFP is not <em>employment — </em>it simply indicates how many adults are either <em>working OR looking for work</em>. The Quarterly Bulletin suggests that about 25% of work-eligible adult men and over 80% of work-eligible adult women were (according to this data) not even looking for work before the pandemic hit.</p><p>Once the pandemic hit in March-April 2020, LFP rates didn’t change too much. For males above 15 years, LFP dropped from 74% to 72% while LFP for women above 15 years dropped from 22% to 20% (I’m rounding up here). Overall, a drop of about 2%, suggesting that many people were still either working or looking for work (more likely the latter, given the situation). Similar figures exist for younger workers between 15 and 29 years of age.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/608/1*gYKhu-2nhoY3qlG2XPKucQ.png" /><figcaption>Snapshot of Statement 1 of <a href="https://www.mospi.gov.in/documents/213904/301563//PLFS_Quarterly_Bulletin_Jul_Sep_2020m1627900970967.pdf/1b035713-c748-12fa-8de7-87da5b98c05c">the PLFS Quarterly Bulletin for July-Sep 2020</a>. Page 2.</figcaption></figure><p>What’s <em>really</em> staggering to me is the change in the Worker Population Ratio (WPR). Between the Jan-March quarter and the April-June Quarter of 2020, the population of males above 15 years <em>actually working</em> dropped from 67% to 57%. This means that if we had (for example) 270 million males actually working in Jan-March 2020, about 40 million of them had stopped working in April-June. For context, imagine the entire population of Mumbai stopping work. Now imagine it a second time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/603/1*I9uiaY8p2LtISWe5e7BzAw.png" /><figcaption>Snapshot of Statement 2 of <a href="https://www.mospi.gov.in/documents/213904/301563//PLFS_Quarterly_Bulletin_Jul_Sep_2020m1627900970967.pdf/1b035713-c748-12fa-8de7-87da5b98c05c">the PLFS Quarterly Bulletin for July-Sep 2020</a>. Page 3.</figcaption></figure><p>The drop for females is slightly better, from about 20% to 16%, but given that female participation in the labour force is so bad to begin with, the outcomes may be more extreme here, in ways that cannot be captured by tabular statistics. It may take a while for us to understand what impact COVID-19 and associated lockdowns had on female work and how much of it recovers over the long run.</p><p>An important point to note here is that I’m using “actually working” in a very loose sense. The Worker Population Ratio uses a criterion called ‘Current Weekly Status’ or CWS to determine if a person is working or not. CWS requires surveyors to ask workers if they have worked for a minimum of one hour in the week preceding the survey.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/798/1*GGn_ylI9rJhWXf2sN-EEug.png" /><figcaption>Snapshot of Current Weekly Status (CWS) definition. From <a href="https://www.mospi.gov.in/documents/213904/301563//PLFS_Quarterly_Bulletin_Jul_Sep_2020m1627900970967.pdf/1b035713-c748-12fa-8de7-87da5b98c05c">the PLFS Quarterly Bulletin for July-Sep 2020</a>. Page 3.</figcaption></figure><p>Now, this definition is deliberately designed to be loose, mostly to capture types of work different from the stereotypical 40-hour work week. However, it can also give us a fairly inflated picture of employment. One, or even 10 hours of work in a week may not mean much in wages, unless you’re in a really elite sector. This raises questions on how meaningful such employment actually is. Many workers during the April-June lockdown may not have earned enough for their families.</p><p>Was there any good news? A little bit. In the following Quarter (July-Sep 2020), WPR rose again a little bit. For male workers above 15 years, WPR rose from 57% to 65%, while for female workers above 15 years, it rose from 16% to 17%. This isn’t <em>great </em>news though. Note that WPRs for males and females in Jan-March was 67% and 20%. So many people who were working in Jan-March did not return to work in July-Sep.</p><p>If (to continue the earlier example) we lost 40 million male workers during the lockdown, we only got about 30 million of them back. In this example, about 10 million hadn’t returned by July-Sep. Imagine the entire population of Bengaluru not stepping out after the lockdown was lifted.</p><p>Again, the picture is worse for women workers above 15 years. If (for example), we had 80 million women workers in Jan-March, that number would have dropped down to 64 million during the lockdown, but only 68 million would have returned to work by July-Sep. In this example, over 12 million workers wouldn’t have returned to work once the lockdown was lifted.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/0*iTMfm6zMwckIkj-c.jpg" /><figcaption>Brickworker in Salem, Tamil Nadu. Source: <a href="https://www.maxpixel.net/Labour-Bricks-Salem-Hiring-Sengal-India-Tamilnadu-2402042">https://www.maxpixel.net/Labour-Bricks-Salem-Hiring-Sengal-India-Tamilnadu-2402042</a></figcaption></figure><p>Several things to keep in mind before ending:</p><ul><li>This data ends at July-September 2020. We’ll have to wait and see for employment data in future quarters to see what the full extent of the recovery is.</li><li>I’ve only written about broad-brush pictures for male and female workers above 15 years of age. The Bulletin contains more data disaggregating workers by industry, employment type and so on. I’d encourage you to take closer looks at these before drawing deep conclusions (I know I’ll be doing the same).</li><li>Please read Sections 3 and 4 of the Bulletin to understand the conditions under which this data was collected.</li><li>A reminder that this is <em>urban</em> data. The data for rural workers may show a completely different picture.</li></ul><p>Lastly, one thing I haven’t summarised here is the actual <em>unemployment </em>rate (how people were unemployed as opposed to employed).</p><p>I’m including a snapshot of the unemployment table below (Statement 5 in the Bulletin, Page 6). Following a long tradition established by annoying maths and statistics teachers, I’m leaving this one as an exercise for the reader.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/718/1*0sIvufXBL2HLcRxREmrG6w.png" /><figcaption>Statement 5 (Unemployment Rate) from <a href="https://www.mospi.gov.in/documents/213904/301563//PLFS_Quarterly_Bulletin_Jul_Sep_2020m1627900970967.pdf/1b035713-c748-12fa-8de7-87da5b98c05c">PLFS Quarterly Bulletin July-Sep 2020</a>. Page 6.</figcaption></figure><p>PS — A plug. If you want another dive into India’s employment data, my colleague at IIHS, Viola Fatima Lewis, <a href="https://www.indiaspend.com/data-viz/where-do-indians-work-764910">has just released an explainer on workers in India, over at IndiaSpend</a>. It’s part of a larger series of explainers on urban data, so check it out.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7400bee134c7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Amogh Arakali’s #FromMyReadings, Issue 6, 2021]]></title>
            <link>https://amogharakali.medium.com/amogh-arakalis-frommyreadings-issue-6-2021-c44b3b2fba6c?source=rss-17a040da3940------2</link>
            <guid isPermaLink="false">https://medium.com/p/c44b3b2fba6c</guid>
            <dc:creator><![CDATA[Amogh Arakali]]></dc:creator>
            <pubDate>Mon, 26 Jul 2021 15:33:49 GMT</pubDate>
            <atom:updated>2021-07-27T06:59:02.587Z</atom:updated>
            <content:encoded><![CDATA[<h3>#FromMyReadings, Issue 6, 2021</h3><h4>On Correcting Misconceptions about Economics</h4><h3>1. On Correcting Popular Misconceptions about Economics (Noah Smith)</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*L_ZAFu8v_uRGS3lW.jpg" /></figure><p>It’s good to see more folks speak out about misconceptions around contemporary economics. <a href="https://noahpinion.substack.com/p/is-economics-an-excuse-for-inaction?fbclid=IwAR1awnsI3X71MOEotAcPQmaZgL50JB_2SUQARcuAoW6oMt4OQCE4gOnNlvs">This piece by Noah Smith</a> echoes points I’ve grown tired of repeating over the last few years:</p><ol><li>That public perceptions of economics is quite different from the way economists themselves frame their discipline.</li><li>That major historical figures in economics (like Adam Smith), often held very nuanced positions on free markets and government intervention. Their works have been misrepresented in today’s public media. <br>(As an aside, I recommend reading the chapters on Smith in <a href="https://scholar.harvard.edu/bfriedman/religion-and-rise-capitalism">Benjamin Friedman’s 2020 book</a>)</li><li>That economic practice has become much more empirical in the last 15–20 years. This isn’t the field of ‘unrealistic models’ that it was often criticised as.</li></ol><p>That being said, here are my caveats to his piece:</p><p>— —</p><p><strong>First Caveat:</strong> <br>Noah argues that poor public perception of economics stems from popular (American) conservative voices hijacking the field and turning it into a “pop version of conservative ideology”. He may be right about this to an extent, especially in a US context.</p><p>But based on my own experiences and interactions, <strong><em>I’d argue that a lot more damage to economics’ image comes from poor teaching, poor course design, and a lack of attention to good pedagogy</em></strong>. This is particularly true when economics is taught as a minor subject or as an optional elective in other programmes.</p><p><a href="https://arakaliamogh.wordpress.com/2015/10/17/another-year-another-nobel-prize-another-rant-about-economics/">I’d ranted about this six years ago</a> (see towards the end of that post). By and large, my position hasn’t changed too much since then (though I’d put it more politely nowadays). I still believe there’s a dearth of good pedagogy in introductory economics courses.</p><p>— —</p><p><strong>Second Caveat:</strong> <br>There’s a lot of (justified) celebration over economics becoming more empirical. However, a naive turn to empiricism is almost as bad as refusing to move beyond theory. More data from the ground doesn’t automatically make better practice. <strong><em>Empirical studies will still need to engage with </em></strong><a href="https://files.eric.ed.gov/fulltext/ED542066.pdf"><strong><em>deductive and inductive models</em></strong></a><strong><em> to improve our understanding of economic phenomena.</em></strong></p><p>A lot also depends on <em>how</em> that data is collected. If economic practice blindly replaces “more deductive models” with “more Randomised Controlled Trials (RCTs)” or “more Panel Data”, we’ll only end up repeating our mistakes, just from a different direction.</p><p>A lot will therefore depend on careful research design, with studies customised to address the contexts of what’s being studied. <strong><em>Good research design is far more important than a turn to empiricism.</em></strong></p><p>— —</p><p><strong>Third Caveat:</strong> <br><strong><em>There’s still a lot of bad practice in economics.</em></strong> For instance, a Randomised Controlled Trial (RCT)in Nairobi last year <a href="https://www.devex.com/news/a-new-research-experiment-in-kenya-raises-questions-about-ethics-98039">involved threatening to cut off water and sanitation services to observe if homeowners would repay debts</a>.</p><p>This kind of unethical work does get criticised more frequently nowadays. However, it is still pretty rampant and needs to be checked.</p><p>— —</p><p><strong>Fourth Caveat:</strong> <br><strong><em>Major economic debates, particularly on theory, are still largely conducted in the west</em></strong>, mostly in the United States and the United Kingdom. They still mostly take place between western universities. There are very few instances of Asian, African, or South American work being reflected in ‘international’ debates on Economics (except when affiliated to a western university or an organisation like the IMF).</p><p>In fact, I suspect this disparity has worsened in recent years, as education systems in parts of the developing world have broken down. Opportunities for academics in countries like India have diminished, forcing them to move to western universities. This has pushed them to participate more in debates and discourse there, with less theory being grounded here.</p><p>When was the last time we saw someone from (even the elite) Indian institutions make a mark in international circles, without being affiliated to a western or international platform? The Azim Premji University in Bengaluru is doing some notable work (see <a href="https://www.core-econ.org/it-works/azim-premji-university-india/">their contributions to the CORE textbook</a>), but that’s a rare bright spot.</p><p>This disparity needs to be corrected, and urgently. It’s not an easy problem to fix. In fact, with a turn towards empiricism and a heavier reliance on RCTs, economic research is becoming too expensive to be conducted by anyone outside the richer institutions of the west (see <a href="https://www.project-syndicate.org/commentary/why-does-the-global-north-dominate-development-economics-by-arvind-subramanian-and-devesh-kapur-2021-03">Subramanian and Kapur on this</a>).</p><p>— —</p><p><strong>Fifth Caveat:</strong> <br>My position on market-government relations hasn’t changed. I still think governments are most effective when they’re building public goods or addressing market failures, but are less effective when they’re trying to replicate the private sector.</p><p>Very simply, I say yes to governments in vaccination drives and no to governments in running airlines (except emergency air services).</p><p>This stance should be pretty obvious to most people nowadays. However, as we move away from privatisation’s sugar rush of the 80s and 90s, there’s a possibility we’ll head in another problematic direction. <strong><em>Sometime in the future, we may start demanding government intervention in places not needed.</em></strong> And governments may be more than happy to oblige.</p><p>Worse, we may deepen the unholy alliance of governments and big businesses, <em>as governments try to intervene more, while also cutting expenditure</em>, contracting out public services to large businesses in return for public resources.</p><p>This is something to watch out for. Economists increasingly favouring government intervention is a good thing, but only because it checks an older bias to privatisation. Their primary stance ought to be that there are no universal solutions to problems, only contextual ones.</p><p><em>PS — Apologies for a long delay. There was a lot happening in my personal and family life over the last few weeks which required my full attention. I’m slowly coming back to my routine.</em></p><p><em>Posts should be more regular in the coming weeks, although I’ll be restricting myself to one reading per post for the next few issues.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c44b3b2fba6c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[On a Break]]></title>
            <link>https://amogharakali.medium.com/on-a-break-a96da95a1c96?source=rss-17a040da3940------2</link>
            <guid isPermaLink="false">https://medium.com/p/a96da95a1c96</guid>
            <dc:creator><![CDATA[Amogh Arakali]]></dc:creator>
            <pubDate>Mon, 28 Jun 2021 15:33:58 GMT</pubDate>
            <atom:updated>2021-06-28T15:33:58.263Z</atom:updated>
            <content:encoded><![CDATA[<p>I’ll be on a short break from blogging. There are some personal matters which require my time for a couple of weeks. I should return soon. Meanwhile, everyone take care!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a96da95a1c96" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Amogh Arakali’s #FromMyReadings, Issue 5, 2021]]></title>
            <link>https://amogharakali.medium.com/amogh-arakalis-frommyreadings-issue-5-2021-875a1df4521c?source=rss-17a040da3940------2</link>
            <guid isPermaLink="false">https://medium.com/p/875a1df4521c</guid>
            <category><![CDATA[informal-economy]]></category>
            <category><![CDATA[maternal-health]]></category>
            <category><![CDATA[macroeconomics]]></category>
            <category><![CDATA[urban]]></category>
            <category><![CDATA[inflation]]></category>
            <dc:creator><![CDATA[Amogh Arakali]]></dc:creator>
            <pubDate>Mon, 14 Jun 2021 15:32:56 GMT</pubDate>
            <atom:updated>2021-06-14T15:32:56.768Z</atom:updated>
            <content:encoded><![CDATA[<h3>#FromMyReadings, Issue 5, 2021</h3><h4>Exploring Threshold Inflation Rates for India; Using the Past to Understand the Present; Informal Work and Maternal Health in Delhi</h4><p><strong><em>[Before you read on, I urge you to </em></strong><a href="https://amogharakali.medium.com/frommyreadings-issue-1-2021-bcfacdf8d978"><strong><em>take a look at the beginning of my earlier post</em></strong></a><strong><em> and donate to those causes if you can. India can do with more COVID-19 relief efforts. Thank you.]</em></strong></p><h3>1. Exploring Threshold Inflation Rates for India (Dholakia et. al)</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/200/0*TnHlGRZBfMoyYMyc.png" /></figure><p>The Reserve Bank of India’s Development Research Group (RBI DRG) <a href="https://rbi.org.in/Scripts/PublicationsView.aspx?id=20360">released a new publication in May,</a> (henceforth Dholakia et. al) which explores the idea of ‘threshold inflation’ in Indian contexts.</p><p>Macroeconomists generally agree that <a href="https://poole.ncsu.edu/news/2019/06/13/inflation-is-healthy-for-the-economy-but-too-much-can-trigger-a-recession/">a small amount of price rise is good for the econom</a>y, but inflation becomes a problem once it crosses a certain limit. This limit is what threshold inflation is supposed to represent — it’s the inflation rate beyond which, price rise becomes detrimental to economic growth.</p><p>The question worth asking is — can an economy have an <em>ideal</em> threshold inflation rate, that policymakers can aim for to increase economic growth? The existence of such a rate can be very useful, as it allows central bankers to plan their targets (especially long-term targets) better. This is what the RBI paper’s authors seek to understand.</p><p>This is a fairly technical paper requiring some knowledge of macroeconomic theory. The authors’ central argument is that a threshold inflation rate can indeed be defined. For any economy (they say), the threshold inflation rate is equal to</p><blockquote>“…the long-run equilibrium rate of inflation that maximises the steady state growth (SSG) within the relevant range of values”.</blockquote><p>There are three points a layperson must note before unpacking that jargon-laden definition.</p><p>First, it’s important to remember that terms like ‘inflation’, ‘economic growth’ and ‘long run’ have multiple definitions within formal economic practice. Each of these definitions is narrow, technical, and meant for specific situations. They come loaded with a history of discourse which distinguish it from similar-sounding terms. Thus, a term like “<a href="https://www.yourarticlelibrary.com/macro-economics/growth-models/steady-state-growth-of-economy-meaning-and-properties/31189">steady state growth</a>” is different from more common ideas of growth, such as year-on-year GDP growth rates (which a layperson is more likely to read about in a newspaper).</p><p>Second, a huge amount of economic practice involves making choices between difficult options (what is usually called a ‘trade-off’). If you have studied Operations Research in business school, you would have already encountered two types of trade-offs — maximisation (increasing the value of something good, given specific constraints) and minimisation (decreasing the impact of something bad, given specific constraints). In this paper, the authors talk about threshold inflation “maximising” steady state growth, subject to certain constraints. For any reason, if these constraints change, then our frameworks, approaches, and conclusions about threshold inflation need to change as well.</p><p>Third, it’s really important to remember that economics is far from being a settled science, particularly in macroeconomics. While the field is dominated by a few schools of thought, there is considerable debate and contention over the validity of basic frameworks and models often used by practitioners. Very often, economic crises (like the 2008 Recession or <a href="https://www.bloomberg.com/opinion/articles/2020-09-16/how-1970s-oil-prices-stagflation-changed-the-u-s-economy">the 1970s American Stagflation</a> trend) can trigger re-evaluation. Therefore, we must keep in mind that the ideas discussed in papers like these are always works in progress, subject to re-evaluation as more knowledge comes to light.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yZVrvbJxCgE6QP2i42T3WA.png" /><figcaption>A table summarising theoretical approaches to threshold inflation. <a href="https://rbi.org.in/Scripts/PublicationsView.aspx?id=20360">From Dholakia et. al, 2021 (The RBI Paper on Threshold Inflation)</a>.</figcaption></figure><p>Threshold inflation for India has been explored before, even within the Reserve Bank. Dholakia et. al has two useful tables — one that documents theoretical, and another that documents empirical investigations into threshold inflation. The authors argue that recent theoretical literature (see Table 1 above) make a case for GDP growth rates that can go both up and down as inflation rates increase (technical term: a ‘<a href="https://www.statology.org/monotonic-relationship/">non-monotonic</a>’ relationship).</p><p>If valid, this implies that there is some point where growth rates reach a maximum level (in relation to inflation rates) before declining. This point would be a threshold level for inflation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZxOLmdvBw2dxcNTCtFlFPw.png" /><figcaption>Empirical investigations into Threshold Inflation for India. Table from Dholakia et. al, 2021 (the RBI Paper on Threshold Inflation)</figcaption></figure><p>If theory suggests that threshold inflation rates are possible, then what is that rate for India? The table on empirical investigations (see table above) provides us with a summary of inquiries into this. As with most macroeconomic studies, empirical investigations are complicated by differences in the availability of data and types of frameworks and methods employed by the scholars. Should one use annual inflation data to calculate estimates of threshold inflation? Monthly? Quarterly? Should we measure <a href="http://www.arthapedia.in/index.php?title=Wholesale_Price_Index_(WPI)">wholesale prices</a> or<a href="http://www.arthapedia.in/index.php?title=Consumer_Price_Index"> retail prices</a>? What kind of econometric regression should investigators employ? Each question is answered differently by different investigators, leading to different conclusions.</p><p>However, this is not a bad thing. On the contrary, if multiple approaches and methods lead to a fairly similar band of results, then that’s usually better than a single framework yielding a single value. The table above suggests that most scholars recommend threshold inflation rates for India to be between 4% and 6%, if we measure wholesale prices. Adding to this literature, Dholakia et. al employ a cross-country comparison approach from which they derive India-specific estimates (this is too technical to explain here— please see <a href="https://rbi.org.in/Scripts/PublicationsView.aspx?id=20360#L11">this section of the paper</a>, if interested).</p><p>Here, they argue that threshold inflation depends on two key factors — the country’s <a href="https://www.investopedia.com/terms/f/fiscaldeficit.asp#:~:text=A%20fiscal%20deficit%20is%20a,spent%20in%20excess%20of%20income.">Fiscal Deficit (FD)</a> and its <a href="https://www.investopedia.com/terms/c/currentaccountdeficit.asp">Current Account Deficit (CAD)</a>, measured as percentages of the GDP.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Zf-eyHaG83WYF6-uzOuRtg.png" /><figcaption>Dholakia et. al’s estimates for threshold inflation based on levels of Fiscal Deficit (FD) and Current Account Deficit (CAD). From <a href="https://rbi.org.in/Scripts/PublicationsView.aspx?id=20360">Dholakia et. al, 2021 (The RBI Threshold Inflation Paper)</a></figcaption></figure><p>Essentially, their estimates of a threshold rate implies an annual consumer price index measuring between 6.08% and 6.15%, depending on fiscal deficit and CAD levels. These estimates are slightly on the higher side compared to most other studies in Table 2 above. Dholakia et. al suggest that higher fiscal deficits, assuming CADs between 2 and 2.5% of the GDP, lead to lower threshold inflation rates, when measuring annual consumer prices. In an ideal scenario, a low fiscal deficit of 5% of GDP, with a CAD of 2% can lead to an inflation rate of 6.15% and a GDP growth rate of 7.74%.</p><p>Dholakia et. al’s claim to speciality however, is in proposing a theoretical model that links fiscal deficit and current account deficit levels to threshold inflation and GDP growth. If more studies in the future validate their claim, it can provide policymakers with a useful framework by which they can estimate the impacts of fiscal and trade policies upon GDP growth rates via inflation. Once again however, please note that more studies and observations are needed before accepting the model as standard practice.</p><p><strong><em>On an off-note:</em></strong> <a href="https://www.reuters.com/world/india/indias-may-wpi-inflation-accelerates-1294-yy-2021-06-14/">India’s Wholesale Price Index (WPI) is touching 13%</a>, well above any threshold level noted in Table 2. However, please note that the WPI is not directly comparable to Dholakia et. al’s study, since the authors use consumer prices (the WPI measures wholesale prices). That being said, a 12.94% Year-on-Year jump in WPI is quite worrying.</p><h3>2. Using the Past to Understand the Present (Ed Glaeser)</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*4wLM4cVkb7-VL_ir" /><figcaption>How much can ancient cities tell us about today’s urbanisation?</figcaption></figure><p>Urban economist Ed Glaeser has released <a href="https://www.nber.org/system/files/working_papers/w28814/w28814.pdf">a new working paper at NBER</a>, where he asks what developing cities today can learn from the cities of the past. He points out that many common urban problems today, such as pollution, traffic congestion, and disease contagion, are not new issues and have affected cities for centuries.</p><p>He proposes using the past to better understand the present and outlines some key lessons from studying cities in history:</p><ol><li><strong><em>Political power shapes drives growth of the world’s largest cities</em></strong> rather than commerce;</li><li>Transportation infrastructure used to be more influential in shaping cities in the past (due to enabling market access), but <strong><em>its power has declined today</em></strong>;</li><li><strong><em>Infrastructure functions best when combined with incentives</em></strong> for adopting sewers but discouraging highways;</li><li>In the west, cities depended on <strong><em>a nexus of property rights that are weak in other developing cities</em></strong> of the world;</li><li><strong><em>Local conditions</em></strong> determine the best choice for institutions that manage infrastructure.</li></ol><p>I can see scholars from other disciplines rolling their eyes about economists saying the same things they’ve been saying for years. This may be particularly true for Lesson 1. That politics shapes space is now fairly well-established among urbanists (see <a href="https://oxford.universitypressscholarship.com/view/10.1093/acprof:oso/9780199270057.001.0001/acprof-9780199270057">Neil Brenner’s work on rescaling of statehood</a> for example), and historians and urban planners have long studied (among other examples) <a href="https://en.wikipedia.org/wiki/Haussmann%27s_renovation_of_Paris#Legacy">Hausmann’s Paris</a> as an example of how big cities can be shaped by political power, often far more than economy or trade.</p><p>However, Glaeser’s paper nevertheless carries value (see <a href="https://amogharakali.medium.com/frommyreadings-issue-2-2021-14c5a4cdb910">point 3 of my earlier post</a>, where I’ve talked about people from other fields entering established domains). Some useful observations early in the paper are on the role of transportation technology. In a style <a href="https://www.jstor.org/stable/2937739">reminiscent of Paul Krugman,</a> he mentions how the invention of motor trucks has allowed goods to be transported directly to bigger cities, bypassing mid-sized towns along the way, thereby leading to a greater concentration of megacities in our current age.</p><p>I won’t go too much into the rest of the paper — it’s a fairly easy read and I strongly recommend laypersons to go through it for themselves.</p><p>However, I do want to say this paper caught my eye for its central argument — of using historical data (data going back centuries) to understand the present. I believe this to be an underrated methodology in both economics and urban/regional studies, with some fascinating uses, yielding interesting insights. What I’ve found particularly useful is its ability to place our present conditions within gigantic timescales, allowing us to appreciate our current predicaments within broader contexts.</p><p>Natural scientists, especially in geology, paleontology, and climate science have mastered this technique, leading to the now-famous framing of our times as<a href="https://anthropocene.info/about.php"> <em>the Anthropocene</em></a>. In recent years, with the rise of interdisciplinary practices, this practice has spread to other fields as well. I was particularly intrigued by a book <a href="https://www.amazon.com/Ten-Thousand-Years-Inequality-Archaeology/dp/0816537747"><em>Ten Thousand Years of Inequality: The Archaeology of Wealth Differences</em></a><em> </em>edited by Dr. Timothy Kohler, an anthropology professor at Washington State University. The book explores research and methods to place current debates on inequality within the context of the entire <a href="https://en.wikipedia.org/wiki/Holocene">Holocene</a> era starting about 12,000 years ago.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/338/0*6UtmYV2Xq95SYyCl.jpg" /></figure><p>I find the book fascinating for both its endeavour to place a contemporary political debate within a larger historical time-scale, as well as the use of archaeological and anthropological methods to understand an economic issue. I can only hope more such methods are explored and refined in the years to come.</p><h3>3. PLUG <em>| Breastfeeding knowledge and practices of working mothers in the informal economy in New Delhi (Antara Rai Chowdhury, Aditi Surie, Gautam Bhan)</em></h3><p>My colleagues at the Indian Institute for Human Settlements have <a href="https://www.sciencedirect.com/science/article/pii/S0277953621004020?dgcid=author">released a paper on a study they conducted before the pandemic</a>, where they surveyed women workers in the informal economy of Delhi with young children.</p><p>The study aimed to understand how the nature of informal work can shape maternal practices like breastfeeding. Among their findings, they discovered that informal workers were more likely to go back to work within six months of birth, leading to early weaning of the children. Furthermore, the lowest paying category of informal work— home-based work — was strongly associated with exclusive breastfeeding. More broadly, the study suggests that the nature of informal work can shape maternal and child health practices. The authors also outline some probable interventions which could work towards improving health conditions for both mothers and children.</p><p>Since this was written by my colleagues, I’m not going to comment on the paper here. I’ll send them any feedback via institutional channels. However, I will point out that these arenas of interdisciplinary research highlight the importance of linkages between work and home, economy and society. Eventually, economic planning and policy design cannot afford to ignore these links.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=875a1df4521c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Amogh Arakali’s #FromMyReadings, Issue 4, 2021]]></title>
            <link>https://amogharakali.medium.com/amogh-arakalis-frommyreadings-issue-4-2021-79011d6ff6ac?source=rss-17a040da3940------2</link>
            <guid isPermaLink="false">https://medium.com/p/79011d6ff6ac</guid>
            <category><![CDATA[pandemic]]></category>
            <category><![CDATA[research]]></category>
            <category><![CDATA[indian-economy]]></category>
            <category><![CDATA[cricket]]></category>
            <category><![CDATA[covid19]]></category>
            <dc:creator><![CDATA[Amogh Arakali]]></dc:creator>
            <pubDate>Mon, 07 Jun 2021 15:32:30 GMT</pubDate>
            <atom:updated>2021-06-07T15:32:30.361Z</atom:updated>
            <content:encoded><![CDATA[<h3>#FromMyReadings, Issue 4, 2021</h3><h4>A Different Type of Long COVID, The Importance of Research That Doesn’t Make Sense, What a Cricket Crisis Says About Our Behaviour in Good Times</h4><p><strong><em>[Before you read on, I urge you to </em></strong><a href="https://amogharakali.medium.com/frommyreadings-issue-1-2021-bcfacdf8d978"><strong><em>take a look at the beginning of my earlier post</em></strong></a><strong><em> and donate to those causes if you can. India can do with more COVID-19 relief efforts. Thank you.]</em></strong></p><h4><strong>1. A Different Type of Long COVID (Zeynep Tufecki’s Insight)</strong></h4><p>I’m becoming a fan of <a href="https://www.theinsight.org/">Zeynep Tufecki’s newsletter <em>Insight</em></a>. Her recent posts have focused on COVID-19, and how our framings of the disease affect the way we think and react to the pandemic.</p><p>Her May 28th post, <a href="https://www.theinsight.org/p/we-need-to-get-real-about-how-the"><em>We Need To Get Real About How the Pandemic Will End</em></a>, was sober, albeit unsurprising. She argues that COVID-19 may follow a similar trajectory to the AIDS epidemic — the majority of deaths may occur <em>after </em>a cure for the disease was found. In our case (she argues), that means the worst of the pandemic may lie ahead of us.</p><p>Tufecki has several reasons for her claim and you can read them <a href="https://www.theinsight.org/p/we-need-to-get-real-about-how-the">here</a>. Essentially, she argues that the majority of the world is likely to get COVID-19 before vaccinations. Less-advantaged regions don’t yet have access to vaccines; the virus is mutating quickly; and regions which have escaped the bulk of the pandemic will find it increasingly hard to remain that way.</p><p>I take two things away from Tufecki’s piece:</p><ol><li>First, we have to stop equating <strong><em>the availability of a solution</em></strong> with<strong><em> the end of a problem</em></strong>. We seem to be prone to this as a species (or perhaps this is just how we are in our times). We think that discovering a medicine means that the disease is on its way out, or that inventing smartphones will automatically lead to more people on the internet. The bulk of a challenge lies in deciding <strong><em>what to do with a solution</em></strong> rather than the solution itself. In the case of vaccines, we still don’t have a system to provide access to the majority of the world.</li><li>This in turn implies the <strong><em>pandemic is unlikely to come to a sudden end any time soon</em></strong>. Dramatic ends may happen in isolated places (Eg: Wuhan in China seems to have recovered fully — see <a href="https://amogharakali.medium.com/frommyreadings-issue-2-2020-d9fb648f79ac">Section 2 of an earlier post</a> about this), but much of the world may have to deal with recurring waves of COVID-19, as the virus mutates and vaccinations don’t keep pace.</li></ol><p>There is a condition called ‘<a href="https://en.wikipedia.org/wiki/Long_COVID">Long COVID</a>’ which has been noted in COVID-19 patients who have seemingly recovered from the disease. COVID-19 recoveries have reported symptoms of fatigue, illness, and more long-term changes to their body that long outlasts discharge from hospital.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fk0MGd9fMbU8Z3MhOE49vA.jpeg" /><figcaption>A Global-Scale Long COVID may not be impossible. Also, I promise my scribbles will get better in future posts. (Illustration: Amogh Arakali, 2021)</figcaption></figure><p><strong><em>We may see a different kind of Long COVID at a much larger scale</em></strong>, where different variants of SARS-Cov-2 continue to mutate in unvaccinated regions and repeatedly return to haunt the planet. This may continue until all parts of the globe build enough capacity to permanently restrict the disease. This is not a prediction, but a possibility. I really hope this sort of Global-Scale Long COVID won’t happen, and that we’ll be able to get much of the world vaccinated soon. But, a Global-Scale Long COVID may not be impossible.</p><p><em>PS — I found this after I had finished writing this post, but a similar opinion on COVID-19 becoming ‘a seasonal’ event in India </em><a href="https://indiaaheadnews.com/india/virus-mutation-is-going-to-be-indias-biggest-challenge-says-indias-top-virologist-26411/"><em>was expressed by CMC-Vellore virologist and microbiology professor, Dr. Gagandeep Kang.</em></a></p><h4>2. The Importance of Research That Doesn’t Make Sense (At Present)</h4><p><a href="https://www.theatlantic.com/science/archive/2021/05/pandemic-virus-mutations-1918-flu/618972/">An <em>Atlantic</em> article contains a fascinating story</a> of how three preserved bodies from the Spanish Flu Pandemic of 1918 helped us better understand the Second Wave of said pandemic.</p><p>The Second Wave hit Europe in late 1918 and (in a pattern now familiar to us) was deadlier and more widespread than the first. During this time, three bodies were preserved in Germany, two boys and a girl. Decades later, tissue samples from their lungs were used to sequence the flu viruses from the Second Wave to help understand how the flu evolved between the two waves to spread more effectively.</p><p>What’s really interesting is how these samples were obtained in the first place. The article says:</p><blockquote>Sébastien Calvignac-Spencer, a virologist at the Robert Koch Institute, in Berlin, came across the samples in this newest study in a stroke of luck. A couple of years ago, he decided to investigate the collections of the Berlin Museum of Medical History of the Charité.</blockquote><blockquote>He wasn’t looking for anything in particular, but he soon stumbled upon several lung specimens from 1918, a year he of course recognized as a notable one for respiratory disease.</blockquote><p>The article also points out why samples from the 1918 flu are so hard to find:</p><blockquote>Doctors then hadn’t even figured out that influenza was caused by a virus. “There’s no way the individual who saved these samples in 1918 had any idea of what could be done to them…”</blockquote><blockquote>[…]</blockquote><blockquote>Without the pathologist who painstakingly preserved these samples and the museums that kept them for decades before science caught up, our understanding of the 1918 flu would be all the poorer.</blockquote><p>Time and again, we come across random, unpredictable patterns of how research takes place.</p><p>Governments and funding agencies tacitly assume that research is linear. Prioritise an area of research first, and then channel funds to that area. This will yield results in that particular field of research within a particular time-frame. Now, this isn’t completely false. <a href="https://www.cfr.org/backgrounder/space-exploration-and-us-competitiveness">Space exploration boomed in the 1960s because the Cold War drove US/USSR funding towards space research</a>. New climate technologies are being developed now because we’ve chosen to prioritise fighting climate change.</p><p>However, it would be a fallacy to assume the opposite — that one should shut down or discontinue funding research that is <em>not </em>an area of priority. Research is hard, much of it is non-linear, and there’s no telling when an obscure field that made no sense (“Why are we funding this?”) will suddenly become an area of importance (or vice versa).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4MKxBbUbiVtQaEDUcR_25A.jpeg" /><figcaption>We can never say when old research will become useful again. At the very least, we should invest in good archives. (Illustration: Amogh Arakali, 2021)</figcaption></figure><p>Another example to illustrate this. In 1985, a Hungarian scientist, Dr. Katherine Karicko, was invited to work in Philadelphia, United States. Her focus area was on the Messenger Ribonucleiac Acid, a single stranded polymeric molecule, that she felt had potential to be developed to treat disease.</p><p>However, funding interest in her area died down and her idea of using this molecule to battle disease was treated with skepticism. According to her story, she lost grant application after application, lost colleagues to funding crunches and was eventually demoted at her university.</p><p>Nevertheless, she kept working on her idea until she and a former colleague figured out how to use a synthetic version of Messenger Ribonucleiac Acid (more commonly known as <a href="https://en.wikipedia.org/wiki/Messenger_RNA">mRNA</a>) to change the human body’s immune response to viruses. <a href="https://www.nytimes.com/2021/04/08/health/coronavirus-mrna-kariko.html">Her work paved the way for mRNA vaccines that are now being developed by Pfizer and Moderna to fight the COVID-19 pandemic</a>.</p><p>As a society, we are doomed to make tough choices on what to fund. Somewhere, a research project will always be defunded because the money is required elsewhere. Nevertheless, researchers live in times where cost-cutting philosophies from certain sectors have bled into areas where they are irrelevant. I’ve heard anecdotes where libraries are asked to present returns on investment projections and laboratories were asked to charge PhD students for facility use.</p><p>We need new approaches to research funding, approaches which acknowledge in their decisions, how research that’s seemingly irrelevant today, may have value later on. These approaches will look and feel less efficient than the ones we currently use, but it’s important to see that economic efficiency is a matter of context and framing. For example, it’s much more efficient to repurpose decades-old research into a quick vaccine, rather than start from scratch. However, such efficiency will not be visible when you start that decades-long project.</p><h4>3. What a Cricket Crisis Tells Us About Our Behaviour in Good Times</h4><p><a href="https://web.archive.org/web/20210530125943/https://www.newindianexpress.com/sport/cricket/2021/may/29/left-in-the-lurch-first-class-umpires-scorers-and-analysts-left-struggling-since-arrival-of-covid-2309044.html">The New Indian Express has an article on how the COVID-19 pandemic is affecting scorers, umpires, and analysts in domestic cricket in India</a>. As with most sports during the pandemic, domestic cricket has been suspended and non-playing staff have been left in the lurch.</p><p>The article mentions how some of this mess was created in the first place:</p><blockquote>“When the number of <em>[domestic]</em> teams increased to 38, the number of matches across age-groups for both genders was more than 2,000.The BCCI <em>[Board of Control for Cricket in India]</em> needed us. And though we were not given contracts, we gave up our jobs for umpiring.</blockquote><blockquote>Last season, the BCCI couldn’t hold matches. We were not wanted anymore. Since we are not part of payrolls, we don’t get any money from BCCI,” says another umpire from the west zone.”</blockquote><p>Officials who are formally employed feel uncomfortable about their colleagues’ situation:</p><blockquote>“When BCCI needed them, they left their jobs and came because private companies didn’t sanction leaves. We are talking about 40–50 days. Last year, the BCCI didn’t need them so it left them in the lurch.”</blockquote><p>From the text above, this situation doesn’t seem to be “caused by the pandemic”. Not really. Instead, the pandemic seems to highlight the structural problems of BCCI’s employment practices. The situation itself seemed to have been built long before COVID-19, with the BCCI not using more prosperous times to secure better working conditions for its domestic cricket staff.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VYL_FV_uPCyjiFq5hTg3gg.jpeg" /></figure><p>The BCCI is hardly alone in this. No one really knows how large the informal economy is in India, but estimates of informal or unorganised work ranges between 85 to 95% of the total workforce, <a href="https://web.archive.org/web/20210530134057/https://www.businesstoday.in/sectors/jobs/labour-law-reforms-no-one-knows-actual-size-india-informal-workforce-not-even-govt/story/364361.html">with the Periodic Labour Force Survey 2017–18 estimating that over 70% of the non-agricultural workforce have no written/enforceable contracts with their employers</a>.</p><p>The lack of enforceable contracts, guaranteeing protections from layoffs or ensuring financial safety during a crisis, leads to much of the workforce being thrown haphazardly into situations like the one above. Popular narratives in Indian media like to focus on the other extreme, of factory strikes by unionised workers who cannot be laid off and the retaliatory lockouts (people still remember the year-long 1982 Bombay Mills strike). However, this does not apply to the vast majority of workers out there, most of whom have little bargaining power to begin with.</p><p>Indeed, this is only one more episode of the pandemic revealing underlying problems in the Indian economy. There have been others. For example, Indian policy long treated rural and urban areas as deeply distinct categories, ignoring the annual migration cycles where workers moved from village to insecure city housing (and back) in search of work. The pandemic threw this open jarringly last year, when workers stranded in poor urban housing during the lockdown were forced to walk back thousands of kilometres to safer housing in their villages.</p><p>Such problems are highlighted during crises, but are not caused during them. Instead, they are the outcomes of policy, political, and social forces building up for decades before the crisis, very often during periods of prosperity. It’s important to remember that not long ago, India was going through a real estate boom in her cities. Many of the construction workers back then had to put up with temporary housing and lack of access to basic necessities, just as they did during the pandemic.</p><p>The seeds of crisis are planted during prosperity. If we don’t use prosperous times to secure ourselves for a rainy day, we will be left drenched. If only some of the money from the real estate boom had been spent on more secure housing or health coverage for construction workers, we may have been able to reduce the immensity of the Great Migrant Walk last year. The same goes for domestic cricket. If only contracts had been more present, more secure and more enforceable, the present crisis could have been averted.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=79011d6ff6ac" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>