Latest news with #WorldDevelopmentIndicators
Indian Express
06-06-2025
- Business
- Indian Express
At World Bank's raised poverty line of $3 a day, extreme poverty rate falls to 5.3% in 2022-23 from 27.1% in 2011-12
With the World Bank raising its threshold poverty line to $3 a day (daily consumption of less than $3) from the earlier $2.15 a day, the extreme poverty rate for India declines sharply to 5.3 per cent in 2022-23 from 27.1 per cent in 2011-12. In absolute terms, people living in extreme poverty fell from 344.47 million to just 75.24 million, latest data from the World Bank shows. At $2.15 daily consumption — the earlier poverty line based on 2017 prices— the share of Indians living in extreme poverty is 2.3 per cent, which is significantly lower than 16.2 per cent in 2011-12, according to the World Bank's estimates. The number of people living below the $2.15-per-day poverty line is recorded at 33.66 million in 2022, down from 205.93 million in 2011. Despite the World Bank revising its extreme poverty line to adjust for global inflation in 2021 prices, India seems to have fared well, with the poverty numbers holding good. At $3 a day threshold, India's extreme poverty rate for 2022-23 rises from 2.3 per cent (at a poverty line of $2.15 a day) to 5.3 per cent, the World Bank has estimates. Adjusting the earlier $2.15-per-day line for domestic inflation from 2017 to 2021, according to sources, brings the threshold poverty line to roughly $2.60—still lower in real terms than the new $3 a day benchmark. The share of Indians living below the revised lower-middle-income category (LMIC) poverty line of $4.20 per day (from $3.65 in 2017 prices) also fell from 57.7 per cent in 2011-12 to 23.9 per cent in 2022-23. In absolute numbers, people living under the revised LMIC poverty line dips from 732.48 million to 342.32 million in a period of 11 years. The World Bank estimates India's population at 1438.07 million in 2023, using its World Development Indicators database and the official Household Consumption Expenditure Survey. Based on the earlier LMIC line of $3.65/ day, India's poverty rate falls from 61.8 per cent to 28.1 per cent, with around 401 million Indians living below the $3.65/day line in 2022. Using the revised LMIC poverty line of $4.20 per day in 2021 prices, India's poverty rate drops to 23.9 per cent from 28.1 per cent in 2022-23. This may seem counter-intuitive, but sources said this is because the new threshold is about 5 per cent lower for India than the inflation-adjusted equivalent of the earlier $3.65 benchmark. When adjusted for domestic inflation between 2017 and 2021, the previous $3.65 line would be roughly $4.40 in 2021 prices, making the revised $4.20 line effectively a lower bar for India. Poverty rates for 2023-24 will be released in October under its Poverty and Inequality Platform (PIP). Sources in the government said these numbers have not been arrived at internally too. Under the $3.65-per-day LMIC line (in 2017 prices), rural poverty fell from 69 per cent in 2011-12 to 32.5 per cent in 2022-23, while urban poverty dropped from 43.5 per cent to 17.2 per cent. The gap by education level was even starker—35.1 per cent of Indians over 16 without any schooling lived below the poverty line in 2022-23, compared to just 14.9 per cent among those with a post-secondary education. According to the World Bank's multidimensional poverty index (MPI), non-monetary poverty in India declined from 53.8 percent in 2005-06 to 15.5 per cent in 2022-23. The index comprises six indicators, namely consumption or income, educational attainment, educational enrolment, drinking water, sanitation, and electricity. The NITI Aayog has estimated that India's population living in multidimensional poverty fell to 11.28 per cent in 2022-23 from 29.17 per cent in 2013-14. Data in the Household Consumption Expenditure Survey (HCES) 2023-24 also indicates a rise in monthly consumption in India. In 2011-12 prices, rural average monthly consumption spending per person increased to Rs 2,079 in 2023-24 from Rs 1,430 in 2011-12, a rise of 45.4 per cent. Urban average monthly consumption expenditure increased by 38 per cent to Rs 3,632 per person from Rs 2,630. Aggam Walia is a Correspondent at The Indian Express, reporting on power, renewables, and mining. His work unpacks intricate ties between corporations, government, and policy, often relying on documents sourced via the RTI Act. Off the beat, he enjoys running through Delhi's parks and forests, walking to places, and cooking pasta. ... Read More
Washington Post
02-05-2025
- Business
- Washington Post
Wild stats about America's place in the world, plus what people actually ask AI
Did you know Romanians lead the world in alcohol consumption? Or that our friends in Singapore and China feel safest walking home at night? We know this thanks to diligent researchers engaged in the important study of global comparisons. But we have to admit: Few activities cause us more pain than wrangling raw international data from the world's approximately 200 countries, depending who's counting, each with its own language, methodology and data quirks. To efficiently and accurately compare nations, you'd ideally have some sort of globe-spanning colossus with experts on the ground helping to collect data everywhere from Angola to Zambia. You'd have, in other words, the World Bank. Thanks to its blockbuster World Development Indicators — long the most popular page on its website — the World Bank is a top purveyor of global data. But for years, much of its deeper data firepower has been hidden, balkanized across one of Earth's largest international institutions. Until now. Last month, the bank's data group dragged us into their D.C. office — shoutout to the security guard who alerted us to an alarming spider in our backpack! — to sneak a peek at Data360, the culmination of a years-long effort to collect, standardize and improve global datasets from the bank and beyond. 'The technology has evolved,' said Haishan Fu, who, as the bank's chief statistician and data officer, runs the team behind Data360. '… To harness data as a transformative force, we need to beef up our capacity.' And, jeez, is it ever beefed up! The bank has already leaped from about 1,500 data series in their central repository to almost 10,000, and they plan to keep adding as additional datasets are found in file drawers and servers in the bank's farther-flung outposts. Or newly collected through the bank's partnerships with member countries, which focus on 'strengthening the data production of developing countries and supporting their national statistics offices,' said Craig Hammer, who helps manage Fu's office and the Data360 launch. The results are mind-blowing. During a wild evening scrolling through 6,417 datasets that include the United States, we kept stumbling across metrics we'd never seen before. Did you know the U.S. is the Saudi Arabia of renewable groundwater, leading the world with 67 percent more than second-place China (somewhat offset by our third-in-the-world freshwater withdrawal rate)? We also lead the world in money sent home by immigrants, with 2.2 times as much annually as the United Arab Emirates or Saudi Arabia, which just about tie for second. And we trail only Norway in the abundance of something called cold-water coral, which feels reassuring even if we don't grasp its importance yet — experts, please write! We also come in sixth in the corals' better-studied warm-water cousins, even before we count territories such as Puerto Rico. America boasts the world's mightiest air-travel and airfreight systems and moves more freight by rail than anyone but China and Russia. But when it comes to passenger rail, we're duking it out with Italy and (prewar) Ukraine for a chance at the top 10. We might rank even lower if the bank still had data for Egypt. We hold the top spot for number of large sports stadiums, ahead of China. But, in a devastating blow to our carnivorous self-conception, we come in only sixth in supply of animal protein per person, behind Iceland, Hong Kong, Mongolia, Ireland and Lithuania. We're among the least vulnerable to food insecurity, and have the 13th-lowest rate of tuberculosis. But our life expectancy has fallen from 14th in 1979 to 62nd in 2022. Meanwhile, our suicide rates have rose dramatically from 2000 to 2019, rising from 66th highest to 25th among men and from 93rd to 29th among women. Fu and her team designed the database with AI in mind. We know chatbot purveyors will slurp up free and open data anyway, so why not improve the odds they'll reproduce it safely and accurately? Their attitude echoes the conventional wisdom about supplying teens with condoms — a topic on which the bank, of course, has data! Speaking of AI, readers often ask us what folks actually discuss with AI chatbots. Now we have an answer. Researchers with the Anthropic Economic Index analyzed a million (anonymous) queries folks typed into the AI assistant known as Claude during February and March and used an in-house AI tool to group them into 630 categories, ranging from 'draft a polite rejection message that maintains professional relationships' to 'role-play as fictional characters in narrative scenarios.' The broad categories tilt toward programming-related questions, an area of expertise for Claude in particular and chatbots in general. (The Post has a content partnership with OpenAI which, according to a news release, means 'ChatGPT will display summaries, quotes, and links to original reporting from The Post in response to relevant questions.') When it comes to individual queries, our favorites were those that asked the bot to write in specific literary genres, allowing us to create a rough popularity ranking. Romantic fiction crosses the line first, followed by magic or supernatural fantasy, science fiction, and humor or satire. What about health-related questions? Hamish Fraser, writing from Brown University, was curious whether people often asked about life-threatening conditions such as heart attacks or sepsis. Our brilliant friends Jeremy B. Merrill and Rachel Lerman produced an excellent analysis last year of an open query dataset that offers more detail. But the most common medical issues people bring to Claude don't strike us as life-threatening. More often, we ask the bots about 'musculoskeletal pain, injuries, and physical rehabilitation.' We're also super curious about 'medication properties and pharmaceutical topics,' as well as 'herbal remedies and traditional medicine.' Folks also came to the chatbot with dental problems and vision issues such as 'eye twitching, light sensitivity, and blurry vision.' In the privacy of that little app or browser window, we also asked about addiction, sleep, dreams, sex, and 'bodily fluids and organ functions.' If we include questions about food and fitness, we see Americans using chatbots for recipes, meal planning and fitness advice (particularly for aging bodies). Mike Malsbary in West Chester, Pennsylvania, asks: 'Who is paying the most attention to artificial intelligence?' He suggests assembly-line workers, librarians — and 'politicians gaming the system.' Killer question, Malsbary! Because we've already dug deep into the jobs with tasks most likely to be performed by AI, we can confidently reply that programmers and developers are the obvious and correct answer to your question. But a new YouGov poll shows that while at least 63 percent of Americans use AI at least occasionally, only about 1 in 10 of us use it for coding. In other words, coders account for a huge share of AI usage but not a huge share of AI users. To cast a wider net, we scanned the most popular types of AI queries for work related questions. After digesting the whole list, we can't shake the feeling that no workplace has been left unturned. We found 'help advance my health care career'; 'explain or analyze international tariffs and trade policies'; and 'create tourism marketing content and business strategy materials.' And yes, even 'provide guidance on industrial manufacturing technologies and processes,' among other queries we can only assume were submitted by factory workers. Folks ask about drafting emails and editing business communications, tossing in some obvious business-related prompts such as 'explain accounting principles and solve bookkeeping problems'; 'analyze financial metrics'; or develop start-up plans. But amid all that tech and business, education stands out. One of the most popular query buckets includes creating teaching materials, while another involves writing or answering multiple-choice questions. Anthropic might dispute this, but the queries paint a disorienting picture of a world where teachers turn to chatbots to churn out educational materials, students press-gang those same bots into answering those same materials, and then teachers go back to the chatbots to create academic scoring systems and rubrics and to 'evaluate and score academic writing.' A separate Anthropic report, which analyzed academic-related conversations from accounts with university email addresses, found that 39 percent of such interactions involved asking chatbots to create or refine essays or classroom materials, while a third of them involved asking the bot to explain or solve problems. The rest of the time, students or professors focused on data analysis, charts, diagrams, research designs and translation. Meanwhile — and this is not legal advice — we're not sure so many people should be going to chatbots to draft, analyze or edit legal documents. Or for legal guidance on marriage and divorce, for that matter. Anthropic's terms of service agree. Greetings. The Department of Data continues its quest for queries! What are you curious about? What's most likely to make us cry? Which non-Roman scripts are most common in the U.S.? Why are suicide rates rising faster in America than in other countries? Just ask! If your question appears in a column, we'll send you an official Department of Data button and ID card. This week, we'll mail them to Hamish Fraser, Mike Malsbary and Paul Tibbits in Bethesda, Maryland.
