Ten interesting things that we read this week

Image: Shutterstock


At Ambit, we spend a lot of time reading articles that cover a wide gamut of topics, including investment analysis, psychology, science, technology, philosophy, etc. We have been sharing our favourite reads with clients under our weekly ‘Ten Interesting Things’ product. Some of the most interesting topics covered in this week’s iteration are related to ‘extended adolescence’, ‘alternative data rush’ ’ and ‘demand management for innovative tech products’.


Here are the ten most interesting pieces that we read this week, ended September 22, 2017.

 1)    How Warren Buffet broke American capitalism  [Financial Times ] 
At 87, Mr. Buffett wields huge influence over the US business and finance, usually positive. He pushed companies to expense stock options, warned of danger in derivatives and taught the public to invest long term in low-cost index funds. But, however much you admire the man, his influence has a dark side - the beating heart of Buffettism, celebrated in a thousand investment books, is to avoid competition and minimise capital investment in the real economy. Recent studies show how exactly those forces — diminished competition, rising profits and lower investment - afflict the US. Corporate mark-ups, a measure linked to profit margins, have increased from 18% in 1980 to 67% today. In a paper, Germán Gutiérrez and Thomas Philippon show how investment has fallen relative to profitability.

While Mr. Buffett did not cause these trends, they are central to his fortune. If he had found a few truly unusual companies and bought them on the cheap there would be no issue. But acolytes are taking his methods economy-wide. Mr. Buffett is completely honest about his desire to reduce competition – something he refers to as “widening the moat”. He tells Berkshire Hathaway managers to widen their moat every year. The Buffett definition of good management is therefore clear. If you have effective competitors, you are doing it wrong. An example is his 1977 purchase of the Buffalo Evening News. He bought this newspaper for $32.5m, a high multiple of its $1.7m operating profit, then launched a Sunday edition and drove the competing Buffalo Courier-Express out of business. By 1986, the renamed Buffalo News was a local monopoly making $35m in pre-tax profit. At the time, it was Mr. Buffett’s largest single investment. His concept of a moat is linked to his views on capital investment: the beauty of one is you do not need the other.

One of his most celebrated purchases is See’s Candies, a company he bought for $25m in 1972. Every year, Mr. Buffett raised prices. So strong was its brand that despite sales growing little, profits grew mightily, with barely any need for capital investment. “The ideal business is one that takes no capital, and yet grows,” he said last year. His statement is unquestionably true for an investor. For an economy, however, it produces the pattern above: low investment relative to higher profits. A line attributed to business partner Charlie Munger in Alice Schroeder’s biography of Mr. Buffett, The Snowball, is revealing: “Munger had always kidded Buffett that his management technique was to take out all the cash from a company and raise prices.” That does sum it up.  If Mr. Buffett in his brilliance had found a few truly unusual companies and bought them on the cheap there would be no issue. But acolytes are taking his methods economy-wide.

These days, Mr. Buffett has two main ways of putting his money to work. On one hand, he is finally investing in physical assets, although only in regulated industries, such as electricity and railroads where returns are largely guaranteed. On the other, he is working with Brazilian private equity firm 3G as it slashes costs to the bone and drives up margins at Burger King and food company Kraft Heinz.  Tax savvy is central to Warren Buffett’s success. Kraft now makes a 23% operating margin and an enormous return on tangible capital. In a competitive market, those high margins ought to present an opportunity for rivals to invest and steal market share. Instead, Kraft competitors, such as Unilever and Nestlé are under pressure from their owners — a mixture of index funds and Buffett-like activists — to match those sky-high margins. If rivals also cut, rather than invest and compete, Kraft can cut even more. A kind of Buffett equilibrium is taking hold.  Mr. Buffett is brilliant at buying into monopoly profits, but he does not start companies or gamble on new ideas.

2)    Extended adolescence: When 25 is the new 18 [Source: Scientific American ] 
It seems that with society’s deepest depravities freely available online, youngsters are growing up quickly: barreling toward adulthood, iPhone in hand, while they SnapChat photos along the way. But new research suggests otherwise. An analysis by researchers at San Diego State University and Bryn Mawr College reports that today’s teenagers are less likely to engage in adult activities like having sex and drinking alcohol than teens from older generations. The analysis used national surveys conducted between 1976 and 2016, which included over eight million 13- to 19-year-olds from varying racial, economic and regional backgrounds. Participants were asked a variety of questions about how they spent their time outside of school and responses were tracked over time. Beyond just a drop in alcohol use and sexual activity, the study authors found that since around 2000, teens have become considerably less likely to drive or have an after-school job and date. By the early 2010s, it also appeared that 12th graders were going out far less frequently than 8th graders did in the 1990s. In 1991, 54% of high schoolers reported having had sex at least once; in 2015, the number was down to 41%. What’s more, the decline in adult activity was consistent across all populations, and not influenced by race, gender or location.

Jean Twenge, professor of psychology at San Diego State and the lead author on the study says that in terms of adult behaviours, 18-year-olds now look like 15-year-olds of the past. The authors initially thought the findings meant teens today are doing more homework or engaged in more extracurricular activities. Yet their data suggest the frequency of these activities has been stable for years, if not on a slight decline. The fact that teens are glued to their computers and smartphones for much of the day may have contributed to the results. Perhaps their socialising and more salacious interests have simply gone digital via texting, sexting and online pornography (today’s teens watch more porn than their predecessors). Yet virtual vice isn’t the whole story because the dip in adult activities began before internet usage became common. The more likely explanation for this new extended adolescence its relationship to affluence. The analysis found adolescents were more likely to take part in adult activities if they came from larger families or those with lower incomes. This mirrors so-called “life history theory,” the idea that exposure to an unpredictable, impoverished environment as a kid leads to faster development whereas children who grow up in a stable environment with more resources tend to have a slower developmental course.

In families with means, there is often more anticipation of years of schooling and career before one necessarily has to “grow up”—there’s plenty of time for that later. The study says that despite growing income disparities, a significant percentage of the US population has on an average become more affluent over the past few decades and is living longer. As a result, people are waiting longer to get married and have children. We’re also seeing a higher parental investment in fewer children.

Some experts caution against reading too much into the new findings, because asking a bunch of teenagers to accurately recount their behaviour has its obvious statistical flaws. But presuming some degree of truth to the new findings, what might postponing adulthood mean for society? The authors feel that instead of pushing young adults to mature faster, we should embrace the cultural shift and develop ways to both meet the psychological needs of modern teens while also setting them up for future success. One such strategy might be expanding mental health services for adolescents, particularly because 75% of major mental illnesses emerge by the mid-20s. She also feels we should stop arbitrarily defining 18 as the age of adulthood and recognise that psychosocial development occurs differently in different people.

3)    A down payment with a catch – ‘You must be an Air BnB host’ [NY Times] 
“If only we had the down payment” may be one of the most frequent complaints among aspiring home buyers in cities across the country. Yifan Zhang, a 29-year-old entrepreneur who often hears this lament among her friends, has come up with a service that tries to help. When she bought a townhouse in Seattle with her husband last summer, she knew that the spare bedroom could generate extra income on Airbnb. But when she learned just how much they could collect each month — enough to cover the mortgage, and sometimes more, her entrepreneurial instincts kicked in: Why not front would-be home buyers money for a down payment, and then collect a share of their Airbnb rental income in return? That was how Loftium, a service in Seattle, came about: It will provide prospective home buyers with up to $50,000 for a down payment, as long as they are willing to continuously list an extra bedroom on Airbnb for one to three years and share most of the income with Loftium over that time.

Loftium expects to appeal to young workers and families who are looking to buy their first home for roughly $600,000 or less. Saving enough money for a down payment has long been one of the biggest obstacles to homeownership. Though consumers generally need to put up at least 20% of the purchase price to avoid paying mortgage insurance, first-time buyers put down far less on average — roughly 8.2%, or about $18,500 — and borrow roughly $207,000. Using Loftium may allow buyers to borrow less or simply get a foot in the door, but it does require an unwavering commitment. As hosts, they must list their extra room year-round, with only eight “freebie” days reserved for their own use — and dozens of strangers are likely to walk through their door for up to three years. Hosts will have the right to cancel up to three guests per year, should they feel uncomfortable with them for any reason. Also, Loftium will determine the size of the down payment it is willing to put up using an algorithm that predicts how much income a room can generate.

The homeowners pay back Loftium through a revenue-sharing agreement in which the company collects roughly two-thirds of the monthly income. If the room isn’t booked nearly enough to generate the expected income, that’s Loftium’s problem, not the homeowner’s. Should a baby arrive or the homeowners want to stop renting the room for any reason before the contract ends, they must pay their share of the nights remaining, plus 15% of that amount, within a week. If the homeowner doesn’t pay what’s owed, Loftium reserves the right to put a second lien against the property, which means the company would be second in line to be paid back (behind the mortgage lender) after the home is sold, refinanced or foreclosed on.

So who is eligible for such an arrangement? Loftium is relying on the mortgage company to vet borrowers’ ability to repay their loans, but it will run its own background checks. Buyers will also need to qualify for a mortgage that meets standards put out by Fannie Mae, which finances nearly a third of all new mortgages in the United States. To start, buyers using the program in Seattle will be able to apply down payments only to mortgages financed by Umpqua Bank, though Loftium said it eventually intended to work with a broader range of lenders (and rental services beyond Airbnb).

4)    Hedge funds see a gold rush in data mining [Source: Financial Times ] 
In recent years, there has been a proliferation of new “alternative data vendors” that trawl through vast pools of digital information and sell it to investment groups desperate for an edge in markets. These vendors often scoop up the digital “exhaust” that people, companies and countries throw out through the normal course of their business and turn it into valuable intelligence.

There are broadly four types of alternative data offerings: a) Website scraping: Many data vendors scrape public websites for information that might be valuable, ranging from downloads of apps and user reviews to what airlines and hotels are receiving bookings through Expedia and Priceline, for instance. Social media sites can also be scraped for hints on consumer views and trends. Companies release vast amounts of data on their websites and even local and national governments are digitising many aspects of their operations. The vendors scrape this “big data” and turn it into tradable signals; b) Credit card tracking: The most valuable seam of data for hedge funds is information that shows directly what consumers are spending money on, and credit card companies are the main gold mine. Although it only offers a partial view of sales trends, combined with other data sets it can offer vital insights; c) Geo-location: Smartphones are equipped with location services that allow us to use map or weather functions, but also let mobile carriers know where we are at any time. That data can be valuable to see what shops, hotels or restaurants we are visiting, a gold mine for hedge funds looking for clues on consumer trends; d) Satellite imagery: In previous decades, an investor might send a junior trainee to a local shopping mall to count how many visitors it gets, or to local farms to check on the health of the latest barley crop. Such information gathering can now be done more comprehensively and automatically by accessing the data of satellite.

There is no suggestion that any hedge fund or alternative data vendor is doing anything illegal by using these new digital data sets. But, lawyers urge caution as some aspects of the business, such as how providers collect the information they sell to hedge funds, and whether it is legal for funds to enter into “exclusivity” agreements with providers, are attracting the interest of regulators and prosecutors concerned that hedge funds are getting an unfair market advantage. The laissez-faire attitude of some of their colleagues and parts of the data vendor community is also under scrutiny. Hedge funds also worry that some vendors do not do a good enough job of scrubbing personally identifiable data from their troves of information. Many therefore have dedicated internal teams that clean the data before it is involved in the investment process.

Hedge funds have good reason to be wary. After a series of arrests from 2010, the industry was upended when dozens of traders, portfolio managers and analysts were prosecuted for using information they gleaned from so-called “expert-network” companies which, like data vendors, tried to match investors with information. The Wall Street traders used company insiders, doctors and others to expand their knowledge of industry trends or companies. The authorities cracked down when traders got their hands on confidential information, such as unpublished drug trial results. To break insider trading rules an individual needs to act on “material, non-public information” received in violation of a duty to keep it confidential. While lawyers say the data sets sold must be material by definition because an investor is willing to pay for them, the vendors typically have permission to sell the aggregate data once a user clicks the box agreeing to terms and conditions. But with few legal precedents and different standards between countries, there is little clarity on the subject.

5)   The lasting importance of human perception in the fleeting world of alternative data [Source: Valuewalk] 
From a pure information acquisition perspective, the enthusiasm surrounding ‘Alt Data’ sources, such as web scraping, geo-locating, weather and satellite imagery, is justified. The opportunity to harness millions of pieces of information collected in real-time as individuals go about their daily online and offline lives is monumental. The information age is transforming into the intelligence age and data analytics is revolutionary for the investment research process, just as it is for so many areas of society. Whilst there is little doubt the industry will continue to accelerate significantly, currently data veracity and the sheer cost of meaningful implementation and in-house analysis are holding many back. Studies have found that only 24% of hedge funds questioned are currently using Big Data in their investment decision-making and 1 in 3 still do not trust it to make decisions on investments.

In the long run, however, datasets will become widely accessible and current ‘Alt Data‘ will not remain alternative. Technology driven, non-financial understanding will play an increasingly prominent part in assessing underlying business value, not just for investment purposes but across many other industries and markets. Standard operating procedures, management, delivery and access improvement, source explosion and cost reduction will lead to the early adopters being caught by the laggards and ultimately the value and proprietary nature of the insight will disappear. The questions asked most frequently will drive the direction of the new datasets and the industry will commoditise, putting it at risk of group think and therefore spotting alpha generating data before anyone else will become increasingly difficult.

Where data is inconclusive or requires corroboration, human-led information acquisition and intelligence analysis provide the opportunity to uncover actionable insight in a highly bespoke and targeted manner. Executed across developed and emerging markets, online and on the ground, situational awareness delivers valuable perspective. It can be as simple as understanding the timeline of a construction project based upon up-to-date images and the input of an engineer, to researching the physical, cross-border supply chain to assess points of interest when there may be possible regulatory or value concern. Mapping networks and understanding board member relationships through online investigative techniques can be particularly useful when planning activism strategies, for example. Furthermore, developing and opaque markets do not provide the same capacity for Big Data analysis; internet penetration and the collection of data is greatly reduced so human based alternative research often remains the most insightful course. Observation of isolated assets (perhaps a mine or offshore platform), difficult to analyse production sites (geographically or politically), or during prolonged projects (noting changes over time) can be invaluable.

Alternative Data is not the silver bullet, much in the same way expert networks or the Bloomberg terminal were not the final say in uncovering proprietary, tradable information. But within the mosaic of alternative investment research, it will be the bespoke, human led intelligence and analysis, alongside ‘Alt Data’, that will continue to look where no-one else is looking and provide answers ahead of the competition.

6)    Why there are never enough new iPhones [Source: WSJ ] 
The big letdown from Apple’s product announcement was the news that its most desirable new gadget, the iPhone X, can’t be pre-ordered until late October and won’t start shipping until November. Even then, analysts are predicting supply shortages. Dealers in Hong Kong expect to sell the phones at a $300 to $400 premium in the first weeks they’re available. Shortages of hot new gadgets are so common, it’s easy to take them for granted. This past year alone, Google’s Pixel phone, Nintendo’s Switch and classic gaming-system reboot, and even Apple’s iPhone 7 Plus have been thin on the ground. You’d think after years of hard-won experience, tech hardware companies could deliver goods when and where there’s demand—especially at Apple, whose chief executive, Tim Cook, made his name as an operations maestro. So, are these shortages some kind of Machiavellian marketing play? Calculated risk management? Or is it the case that manufacturing millions of complex electronic gadgets, and distributing them globally, is really hard? It’s likely a bit of all of these.

Companies often do force early adopters to wait. This is because of the increasingly important pre-order period, says Asokan Ashok, who was at Samsung Research America from 2009 to 2015. He says, the pre-order mechanism, where customers signal their intent to buy a product before it starts shipping, provides early data that is essential to predicting demand for a gadget and distribution of demand across its various configurations—both notoriously difficult to forecast. The calculus gets trickier when the device’s price hits a new high and the company has no precedent to gauge demand. That’s certainly the case with the $999 iPhone X. Apple has “never gone there before,” he says.

Apple may well be facing its biggest demand-forecasting challenge since the original iPhone. While Apple has empirical data on demand for new iPhones, it’s harder to know what proportion of orders will be for the iPhone X versus its somewhat less pricey new siblings, the iPhone 8 and 8 Plus. In addition, many analysts are predicting this could be a “super cycle” of upgraders who have been holding out for a complete redesign—leading to a sales surge like the one caused by the 2014 launch of the iPhone 6. One reason forecasting demand matters so much is the rise in just-in-time manufacturing. Producing millions of phones means coordinating a symphony of component manufacturers, all delivering parts on demand. These partners can’t risk unsold inventory eating into their already razor-thin profit margins. The result is that electronics tend to ship directly from factory to retailer.

One of the biggest challenges top-tier manufacturers face in delivering product on time is their own exacting standards, says Mr. Ashok. Apple is notorious for changing small but important details of its final hardware designs more or less at the last minute. That perfectionism isn’t unwarranted, since hardware can’t be revised once it’s in the hands of the consumer, and the cost of recalls, such as Samsung’s Galaxy Note 7 battery fire fiasco, can be enormous. Even relocating a single screw can have consequences for the dimensions, placement and performance of dozens of other components. Details like this can be worked out in advance given enough time, but in the cutthroat mobile industry, time is something even Apple doesn’t have. A yearly refresh of its phone line leaves only a few months in each design cycle to bring a device out of the lab and onto the factory floor.

When a company brings in new tech—such as the iPhone X’s OLED display, not seen in any previous iPhone—the headaches multiply. Even Apple can’t get around what may be the most critical bottleneck for the iPhone X. At present, analysts say, only one conglomerate can manufacture that OLED screen—and it is Apple’s biggest smartphone competitor. Granted, Samsung Electronics’ visual display business is separate from its mobile business, but Apple appears to be trying to bring in other partners to make OLED displays, including Japan Display and LG Display. Manufacturing smartphones is so complicated, it’s a minor miracle they get to us at all. Companies control whatever they can by pre-announcing products, collecting pre-orders, stoking demand and when necessary managing expectations. They even ponder how much scarcity may increase desire—so that when the product does arrive, weeks or even months later, it feels like it was worth the wait.

7)    The concept of Schizophrenia is coming to end [Source: impactlab.net ]
The concept of schizophrenia is dying. Harried, for decades, by psychology, it now appears to have been fatally wounded by psychiatry, the very profession that once sustained it. Today, having a diagnosis of schizophrenia is associated with a life-expectancy reduction of nearly two decades. By some criteria, only one in seven people recover. Despite heralded advances in treatments, staggeringly, the proportion of people who recover hasn’t increased over time. Part of the problem turns out to be the concept of schizophrenia itself. Arguments that schizophrenia is a distinct disease have been “fatally undermined”. Just as we now have the concept of autism spectrum disorder, psychosis (typically characterised by distressing hallucinations, delusions, and confused thoughts) is also argued to exist along a continuum and in degrees. Schizophrenia is the severe end of a spectrum or continuum of experiences.

Another problem is that schizophrenia is portrayed as a “hopeless chronic brain disease”. As a result, some people given this diagnosis, and some parents, have been told cancer would have been preferable, as it would be easier to cure. Yet this view of schizophrenia is only possible by excluding people who do have positive outcomes. For example, some who recover are effectively told that “it mustn’t have been schizophrenia after all”. Research is now exploring the different ways people may end up with many of the experiences deemed characteristic of schizophrenia: hallucinations, delusions, disorganised thinking and behaviour, apathy and flat emotion.

Indeed, one past error has been to mistake a path for the path or, more commonly, to mistake a back road for a motorway. For example, based on their work on the parasite Toxoplasma gondii, which is transmitted to humans via cats, researchers have argued that “the most important etiological agent [cause of schizophrenia] may turn out to be a contagious cat”. While evidence does suggest that exposure to Toxoplasma gondii when young can increase the odds of someone being diagnosed with schizophrenia, the size of this effect involves less than a twofold increase in the odds of someone being diagnosed with schizophrenia. This is, at best, comparable to other risk factors. For example, suffering childhood adversity, using cannabis, and having childhood viral infections of the central nervous system, all increase the odds of someone being diagnosed with a psychotic disorder (such as schizophrenia) by around two to threefold. More nuanced analyses reveal much higher numbers. Compared with non-cannabis users, the daily use of high-potency, skunk-like cannabis is associated with a fivefold increase in the odds of someone developing psychosis. Compared with someone who has not suffered trauma, those who have suffered five different types of trauma (including sexual and physical abuse) see their odds of developing psychosis increase more than fiftyfold. Other routes to “schizophrenia” involve deletion of a small stretch of DNA on chromosome 22 and inflammation of the brain caused by autoimmune disorders.

All the factors above could lead to similar experiences, which we in our infancy have put into a bucket called schizophrenia. One person’s experiences may result from a brain disorder with a strong genetic basis while another person’s experiences may be due to a complex post-traumatic reaction. Such internal and external factors could also work in combination. Either way, it turns out that the two extreme camps in the schizophrenia wars – those who view it as a genetically-based neurodevelopmental disorder and those who view it as a response to psychosocial factors, such as adversity – both had parts of the puzzle.

Unlike the case for medical conditions like diabetes and hypertension, there is emerging evidence that different routes to experiences currently deemed indicative of schizophrenia may need different treatments. Preliminary evidence suggests that people with a history of childhood trauma who are diagnosed with schizophrenia are less likely to be helped by antipsychotic drugs. It has also been suggested that if some cases of schizophrenia are actually a form of autoimmune encephalitis, then the most effective treatment could be immunotherapy (such as corticosteroids) and plasma exchange (washing of the blood).

8)    China is leaving Donald Trump’s America behind [Source: Financial Times ] 
Michael Moritiz of Sequoia says that the benefits of immigration, the quest for fresh discoveries, the desire for education, the recognition of the benefits of stability, purpose and enterprise are flourishing in China at the very time that they are being maligned, belittled or ignored in the US by Donald Trump.

Take immigrants. The Chinese Government has decided to expand a programme that allows qualified foreign graduates to obtain work and residency permits. It also floated the possibility of expanding the programme to provide for permanent residency. Compare this with the demonisation of Muslims, once welcomed in America, the reversal of the “Dreamers” programme and the dragnets of Homeland Security personnel rounding up illegal immigrants in the US.

Look at education. In China, the central and provincial governments are rushing to build thousands of new schools in rural areas. The thirst for education accounts for a disproportionate amount of household spending. You have only to look at the burgeoning after-school tuition market to see the consequences. Private tutors, who are compensated according to bonus systems, earn an average of about $50,000 a year with a handful of outliers making as much as $300,000.

Then consider factories. While Mr. Trump wants to restore the manufacturing jobs of the 1950s, the Chinese are taking the opposite track. Instead of placing more people on assembly lines, the government wants to install millions of robots over the coming decade. It has set itself the more audacious challenge of raising literacy levels rather than pretending it is possible to return to the past. There are plenty of other examples of how China is advancing while most of the USA is either stuck in neutral or going into reverse. This week several Chinese airlines reported that their earnings were being hurt by travellers using freshly built high-speed train services. A number of local governments are saying that they intend to develop Hyperloop systems.

Meanwhile, China’s army of entrepreneurs is facing the future with an unrivalled sense of adventure and curiosity. The millions of internet-connected bikes that have appeared on the streets of China’s major cities during the past 18 months, although greeted with fury in some quarters, are the most visible emblem of this. Then there is DJI, a Shenzhen company and offshoot of the robotics lab of the Hong Kong University of Science and Technology, that makes 70% of all drones sold around the world. The most impressive example of the Chinese global leadership is in electronic payments thanks to Alipay and WeChat Pay, the payment systems of the country’s two largest internet companies, Alibaba and Tencent. Every day, 600m payments course through WeChat Pay, allowing Chinese millennials to go for months without having to use cash.

Westerners often complain about the policies of the Chinese Government. But from Beijing the world looks very different. Today’s political chaos in Washington and London leaves many Chinese wondering whether their long-held predictions about the eventual collapse of democracy are coming true.

9)    Parrot mimics owner and orders gift boxes from Amazon’s Alexa  [Source: International Business Times ]
A pet parrot went on a bit of an online shopping spree and ended up getting gift boxes for its owner by mimicking its owner and tricking Amazon's voice enabled assistant- Alexa. Corinne Pretorius was surprised when a £10 order for gift boxes arrived at her home and quickly figured out that neither her 45-year-old husband Jan nor her eight-year-old son Jaden had ordered them.

It turned out that the family's African Grey Parrot, Buddy, had mimicked her voice and made the order on her behalf. "I couldn't believe it when I realised that it was Buddy who had used Alexa to make an Amazon order," said Pretorius. "None of us even knew how to use her for that." She continued: "We've had the Echo for about four months and I use it to play music or make to do lists but I've never ordered anything online. On Sunday, we had popped out of the house for a couple of hours, but when I came home, I could hear Buddy talking but couldn't quite make out what he was saying," she said. "Then I heard Alexa say 'sorry I didn't quite get that'. Buddy then said, 'Alexa' and some gibberish, and Alexa replied, 'What is it you want to order?'"

The 39-year-old said she did not think much of the exchange until she received a notification that her Amazon order had been placed. "I hadn't ordered anything and couldn't figure out what had been ordered," she said. "I asked my husband and my son who didn't know what I was talking about so then I asked Alexa: 'What was my last order?' and she said it was these golden gift boxes.

The Amazon spokesman said that customers using Alexa to make voice purchasing must confirm the order by saying "yes". Users can also turn off voice purchasing or require a confirmation code before each order. "Additionally, orders placed with Alexa for physical products are eligible for free returns," he said.

10)    Climate change means ruin to South Asia’s poor; India could make it worse [Source: The Wire ] 
Every monsoon the rains unleash landslides in the Himalayan foothills and floods in the plains, sweeping away families in a fury of loose mud and water. The poorest of the poor are always the worst affected. Their earth and bamboo houses are swallowed by the water as the people try to save whatever they can. The roads become a refuge for people and cattle. Blue and orange tarps appear for shelter, rigged to trees and bamboo poles. Slightly putrid rice is dried on the tarmac, along with clothes and mattresses. Then the wait for relief begins. Everyone scrambles when it arrives, and there is never enough. A few weeks later, the nation forgets the floods. The villagers eventually clear the mud from their homes and wipe them clean. Then the men of each family go to find work as labourers in the Gulf. The cycle of poverty and migration keeps the country going. This year South Asia saw unprecedented weather. The clouds over Assam in India and along the foothills of Nepal just stayed put and dumped a record amount of water, day after day. In Nepal, 159 people have died so far. Twenty-eight are still missing. Crops are ruined and a quarter of a million homes damaged or destroyed. The water flowed south to Bihar and Bangladesh. A separate weather system has brought disaster to Mumbai. More than 1,200 have died across South Asia in all.

A coincidence of timing makes it seem obvious that these unusually severe weather events are a global phenomenon. For years, people in South Asia have been hearing about how climate change will affect them. Apocalyptic images of people suffering alternate drought and deluge have been etched in their minds. They’ve been made to believe that the global order of rich and poor will remain because poorer countries like theirs will be especially cursed by global warming. Each year brings a once-in-a-decade, a once-in-a-century, a once-in-five-hundred-years storm. Human-made changes to weather patterns, mixed with urban development and bad engineering in vulnerable areas, have now left Florida, Texas, Bangladesh and Nepal in the same position. We’re also seeing that the West can vote for a senator who denies climate change but they can’t deny six feet of standing water in the kitchen. Everyone is in the same boat. And we’re also seeing how the politics and planning of this new normal can play out and how climate disasters are a harbinger of social conflict.

In July, people from the village of Tilathi in Saptari district, Nepal, had a feud with their neighbours in Bihar. The Indian Government had built an embankment to protect its citizens, which meant that villagers in Nepal would be inundated every monsoon. When the floods breached the embankment in August, villagers in Nepal refused to let the Indian Government rebuild it. “Our homes and land get submerged because the path of the river has been blocked,” Dev Narayan Yadav, a local teacher, said. The waterways have formed new courses. “It’s flowing backwards now. This is not good.” The Indian media claimed that Nepal had deliberately released flood water to shift the curse downstream. However, the truth is that all the embankments and barrages along the border are built and controlled by India. Nepal has no control. For 25 years, forests have been cut, roads have been built and wetlands encroached. And drainage has been completely disregarded. All across Nepal’s plains, there are road embankments swept away by the force of water escaping submerged villages.

In the town of Gaur near the India-Nepal border, people were stranded for three days. The Indian Government had built an embankment to protect the town of Bairgania, in Bihar’s Sitamarhi district. “We watched the town across the border, which was dry and bright, as we were drenched night and day,” Yogendra Ray Yadav said. The water eventually escaped after a section of the embankment collapsed. The story of the floods moved south.