Sample Participant-Hired Worker Expense Report: Fill & Download for Free

tl;dr the best evidence says that peer-to-peer marketplaces tend to:disproportionately benefit consumers over providers.benefit part-time upper-class workers only — full-time poor workers aren’t really all that better off.not work universally across industries, with wide variability in income on different platforms.preserve existing systemic class-based, age-based and racial discrimination, often with no resolution.have trust mechanisms whose efficacy in keeping the marketplace reputable and safe for consumers is unknown.With this evidence, I am forced to conclude peer-to-peer marketplaces cannot and should not replace a traditional economy (edit: unless these issues are resolved, of course).Below, I describe why this is the case.DefinitionsTrying to figure out what “sustainable” and “uberizing” connotes in practice is hard — as the European Commission’s technical report on the topic succinctly puts it[1]:There is no ‘shared’ consensus on what activities comprise the ‘sharing economy’.A handy diagram for why such definition is difficult:The many kinds of economies, evaluated on axes of for-profit (NFP/FP) and peer-to-peer selling (P2P/B2C).In the diagram above, true sharing and P2P sharing are what people usually can mean by “sharing economies”, but are clearly very different: public bike sharing programs in Europe[2]or distributed file sharing services such as torrenting[3] have different economic dynamics than, say, AirBnb or Uber.To add to the confusion, even within the P2P sharing space, there is a wide distribution of business models to pick from: operating as a pure reseller, a pure multi-seller platform, or somewhere in the middle.Answer StructureI had suspected, prior to researching this question, that it might be too broad and murky to answer conclusively.Thus, my answer will not attempt to define either “sharing everything” or “sustainable” unambiguously.Instead, it will illustrate the space of dimensions in which these terms can be understood, and provide the best research we have right now on all of them:Trust and reputation: Can participants in sharing economies trust what they get to be safe and effective without government regulation?Incentive to participate: When are peer-to-peer marketplaces or true sharing models more attractive to sell in compared to an ordinary market?Labour security and equity: Do peer-to-peer marketplaces by themselves provide reasonable job security and income for their participants?Trust and Reputation… one might be surprised that some Internet markets work at all. When eBay started, it was not obvious that people would send money to nearly anonymous sellers or that these sellers would reciprocate by sending the promised items. Similarly, one might doubt that people would repay peer-to-peer loans, hire babysitters on the strength of a few online reviews, or rent rooms in their house to lightly vetted strangers. Yet all of these transactions seem to be workable. Apart from general goodwill, what are the mechanisms that make this possible?— Einev et. al., 2015[4]Shared economy models (specifially, peer-to-peer marketplaces) implicitly assume that people selling and buying goods are reliable, safe and responsible.In cases where buyers might need additional assurance, platforms often provide a range of mechanisms such as reputation systems, insurance, pre-screening, user education and secure payments.[5]These assurances contribute to a feeling of security around use of some peer-to-peer marketplaces[6], but experts note that these efforts aren’t always reliable (same source as previous footnote):… policy makers need more evidence and analysis to determine how effective these mechanisms are in achieving consumer protection outcomes. With respect to reputation and ratings mechanisms, there are, for example, questions about the accuracy and objectivity of some reviews, as well as the selection or presentation of reviews by peer platforms. Ratings can be false, biased or reflect socially desirable behaviour or strategic manipulation[7]. …Similarly, there is evidence that pre-screening mechanisms in place may not always be of high standard, and consumers may be misled about the nature and quality of these screenings (Hui, 2014)[8].Worse, these trust mechanisms can actively harm end users in unexpected ways:Trust and reputation systems also can invite discrimination. For example, it has been shown that the posting of personal profiles and pictures facilitates discrimination based on the seller’s gender, age, and race (Edelman & Luca, 2014[9]). The reliance on reputation systems and peer-ratings can also invite attempts to harm the reputation of others wilfully, or even blackmail, with a corresponding loss of social capital (Solove, 2007[10]).Privacy issues are another example: Uber and Lyft allow peer consumers to map the progress of their rides to enable them to verify that drivers took the shortest route. This feature is only possible through the systematic collection of location data. More generally, reputation systems generate considerable personal data that may have a high impact on peers and raise difficult privacy issues.Despite all this, it is unknown if trust systems are failing to work enough to require policy intervention. Few empirical studies exist on the efficacy of trust systems — of the ones that do, most empirical studies have focused on discerning bias in reviews, such as Lauterbach et. al’s paper on collusion in Couchsurfing reviews[11]. Research on review bias can become outdated as platforms update their systems to detect dishonest reviews more efficiently and systematically[12] .Policymakers have instead approached safety and reliability in peer-to-peer platforms from a legal, rather than empirical, perspective — a task that is no small feat. Consumer protection law and product liability regulations in many countries applies to sellers of goods, but peer-to-peer platforms, wih their diffusion of responsibility, confound these laws.Experts agree that any regulation must be tailored to different peer-to-peer business models (depending on how “decentralized” the marketplace is)[13], while some argue that existing consumer protection laws are outdated and ineffective in the era of the Internet owing to regulatory capture[14]. An example of where a marketplace has been held responsible for defective goods by a peer seller is the case of Oberdorf vs. Amazon[15], so there is precedent for platforms being treated in this way.True sharing models such as BitTorrent, Napster, bike-sharing programs or energy microgrids have all received widely different regulatory treatment and focus — rather than protecting the consumer, the pimary legal focus tends to be prevention of use for illegal activity. Few avenues, if any, exist to protect the consumer in any of these cases — thus, one may conclusively say that true sharing models have no inbuilt consumer protection.Conclusion: it is too early to tell whether peer-to-peer marketplaces or true sharing models are operating as safely as regulated traditional marketplaces are. Legal issues in determining liability vary from country to country, and (to the best of my knowledge) no universal consensus exists on exactly the right model to proceed for consumer protection or regulation in this space. True sharing models have no guarantee of reliability of goods — you get what you pay for (“free”).Incentive to participateIn the hotel industry, demand can vary widely between peak and offpeak periods. Building hotel rooms is an efficient way to serve a fixed number of consumers, but variation in demand means either periods of very high prices or alternatively empty rooms. A similar story can be told about the taxi industry, where instead of prices adjusting to clear the market, drivers or riders tend to end up queueing or waiting around. In these settings, there is an economic logic to having an elastic supply of flexible drivers or accommodations— Einav et. al. 2015[16]People who participate in a shared economy model tend to be younger, ethnic majorities, comparatively richer, more educated, employed, and living in urban environments.[17].Over 25% of college-graduated Americans with incomes > $75,000 have used a home-sharing service, whereas only 4% of high-school graduates (or less) with an income of less than $30,000 have done so. [18]Employees and self-employed individuals are on average 10–15% more likely to use sharing economy platforms than manual laborers or unemployed persons.[19]Some are in it for money and convenience, some are in it for a sense of community and the fun of it, some are in it because they think their platforms can help them do public good such as reduce waste. [20]This varies across platforms: while not for-profit platforms have optimistic members in it to do social good[21], many in for-profit platforms reject the notion that they’re participating in a sharing economy, preferring instead to state that they’re in a lending or selling economy.[22]Almost universally, “consumers” outnumber “providers” on peer-to-peer marketplaces[23]:Only 1% of all adults have been providers in a ride-sharing model, while 8% of all adults have been participants1.4% of the US population has provided a home in a home-sharing model, while 6% have been participants17% of respondents to Eurobarometer surveys on the subject class themselves as “generic users”, while 5% class themselves as “providers”Cultural attitudes, values and norms impact the degree of participation one sees in “provider” and “consumer” behaviour across countries on shared economy platforms. (The literature on this is too fragmented to summarize — the last citation has enough links on the subject for the interested reader).Coming to the economic side, we have Einav et. al. 2016 (quoted numerous times in the above), who model incentives to participate strictly in terms of economics: peer sellers tend to be more lucrative in markets with low barrier to entry, low costs to advertise, and high demand variability compared to regular “established” businesses.Conclusion: The current basis of participation in shared economies is highly skewed, both in terms of a small minority doing the bulk of providing for the majority and in terms of socioeconomic demographics. Combining both economic incentives and this picture, it seems safe to assume that the impacts of the sharing economy does not reach rural areas as much.Finally, economically being a peer seller rather than a full-time distributor works better only in a limited sample of markets with specific conditions.Labour security and equityAs one of the increasing number of controversies surrounding the ‘sharing economy’ in the period 2014-2015, the dispute over eroding labour security and inequalities is probably the most heated.— Joint Resarch Centre, European Commission[24]Many peer-to-peer marketplaces argue that they are providing opportunity for people who are struggling economically[25].However, a lot of evidence points to the reductive nature of this claim:While Hall and Kreuger reported that Uber drivers earned more than landed professionals, had more flexible schedules, and earn extra money on top of other sources of employment based on Uber data[26], this data did not properly take into account the costs of being an Uber driver. After these costs are factored, compensation for drivers is only just above the minimum wage on aggregate. [27]Surveys have found that income rises for the upper half of the bottom 80% of the population (the so-called “white collar” class), but that in turn comes as a “crowding-out effect” on the bottom half of the bottom 80% by taking on tasks typically associated with “blue-collar” or “pink-collar” roles[28]. Thus, income only rises for some sections of society at the expense of others.Earnings and earning inequality vary widely across platforms — AirBnb providers make as much as $9,000 (median) and $11,000 (mean), whereas TaskRabbiters only make about $2,500 (median) and $6,800 (mean) in comparison[29].Low-income people not only have less desirable assets to bring in to sharing models in the first place, they also feel that they could make more sustainable income outside the platform.I mean like there are many times that you do this and you think, I’d be way better off working at McDonalds because I’d make the same amount of money and I’d have free fries…Working for TaskRabbit is just a fantastic way to always stay at the poverty level, right? But at least you can pay your phone bill and you can buy some food and the landlord isn’t upset with you.— interview with a TaskRabbiter, older man who’d lost his job[30]At the same time, experts note that there is some difficulty in understanding the contribution of sharing economies to the erosion of worker security in general. [31]Larger background forces — such as decline in real wages, high unemployment, and the declining share of labour in national income — are prevalent in prior work, so one should proceed with care when identifying causative factors in the sharing model using existing data.One topic both raised and yet curiously absent in depth on this subject is labour benefits. Most sharing platforms are able to get away without providing any of their providers benefits such as health insurance — if insurance is provided, it is often to the customer, rather than to the provider. Not being able to find definitive reviews on this subject, I have declined to dig into it — in any case, the impact of such benefits is only material in minority countries without a stable social safety net (such as the United States).Conclusion: it is safe to say that, as far as commercial peer-to-peer marketplaces go, it benefits only a certain class of people disproportionately while not being a healthy source of full-time income.Miscellaneous EffectsThere are many other aspects of this phenomenon worth looking at:The effect of sharing economies on social cohesion between members of the community. Studies of this effect have been largely inconclusive or negative:[32]Relevant gaps in the literature exist on the social outcomes of participation, in particular whether and how participating in sharing practices produce social capital, trust, or reputation. At the micro level, there is some (small) evidence of its production within homophilic networks (Schor et al., 2015). On a more macro scale, there is some evidence of its disruption as a consequence of the dis-intermediating effect of algorithms (Parigi & State, 2014). Still at the macro level, there is no relevant research on incremental (or detrimental) effects of sharing economy platforms on social capital in a given area.In the case of trust, despite some claims in popular literature, at the micro level we can also find a definite lack of evidence on the positive effect of participation on interpersonal trust between users (cf. Pais & Arcidiacono, 2016 for an exception). On a more 'macro level', trust building has been conceptualized as institutional trust towards algorithms and brands/companies (Botsman & Rogers, 2010). As it happens with social capital and still at the macro level, there is no evidence, neither positive nor negative, on the effect of sharing platforms on trust within a given geographical area.The market efficiency of sharing platforms compared to existing ones[33]. Most proponents argue that sharing platforms reduce information asymmetry and lead to fairer prices; libertarians see tax savings in avoiding regulation-heavy distribution techniques.Recent developments, such as pushes among platform workers to unionize [34]or even collectively own platforms, may alter how these marketplaces work in income distribution.Structural or novel pricing and discovery mechanisms within peer-to-peer marketplaces. Many platforms have discovered, for example, auction models don’t work when purchasing labour (such as TaskRabbit); at the same time, ensuring fairness in pricing, minimizing “cheaters” who try to game the system, and connecting users to better sources is a moving target.One area of weakness in my analysis is that I’ve focused disproportionately on peer-to-peer marketplaces. Offline sharing schemes, free large-scale crowd-run platforms, and public community efforts have been largely neglected. It’s possible that many of the conclusions above might be reversed if we consider these programs at scale. At best, I can only say right now that there is unprotected risk of participation in these communities as well as that people likely to participate do so out of a sense of altruism, hedonism and utilitarianism.Footnotes[1] https://ec.europa.eu/jrc/sites/jrcsh/files/JRC100369.pdf[2] Bike Share: A Synthesis of the Literature[3] http://krepublishers.com/02-Journals/T-Anth/Anth-18-0-000-14-Web/Anth-18-1-000-14-Abst-PDF/T-ANTH-18-1-007-14-1106-Belk-Russ/T-ANTH-18-1-007-14-1106-Belk-Russ-Tx%5B2%5D.pdf[4] https://web.stanford.edu/~leinav/pubs/AR2016.pdf[5] https://unctad.org/meetings/en/Contribution/dtl-eWeek2017c05-oecd_en.pdf[6] https://core.ac.uk/download/pdf/77240169.pdf[7] A First Look at Online Reputation on Airbnb, Where Every Stay is Above Average[8] Inside Toronto's Uber investigation[9] Digital Discrimination: The Case of Airbnb.com[10] The Future of Reputation: Gossip, Rumor, and Privacy on the Internet[11] http://snap.stanford.edu/class/cs224w-readings/lauterbach09trust.pdf[12] https://arxiv.org/pdf/1710.10061.pdf[13] http://www.oecd.org/going-digital/topics/digital-consumers/peer-platform-markets-workshop/protecting-consumers-in-peer-platform-markets-workshop-summary.pdf[14] The Sharing Economy and Consumer Protection Regulation: The Case for Policy Change[15] When Product Liability Meets the Platform Economy: A European Perspective on Oberdorf v. Amazon[16] https://web.stanford.edu/~leinav/pubs/AR2016.pdf[17] Participation in the Sharing Economy[18] The New Digital Economy: Shared, Collaborative and On Demand[19] Flash Eurobarometer 438: The use of collaborative platforms - ecodp.common.ckan.site_title[20] Participation in the Sharing Economy[21] Domesticating the market: moral exchange and the sharing economy[22] Sharing economy workers: selling, not sharing[23] https://www.pwc.com/us/en/industry/entertainment-media/publications/consumer-intelligence-series/assets/pwc-cis-sharing-economy.pdf[24] https://ec.europa.eu/jrc/sites/jrcsh/files/JRC100369.pdf[25] Does the sharing economy increase inequality within the eighty percent?: findings from a qualitative study of platform providers[26] An Analysis of the Labor Market for Uber's Driver-Partners in the United States[27] https://ec.europa.eu/jrc/sites/jrcsh/files/JRC100369.pdf[28] Does the sharing economy increase inequality within the eighty percent?: findings from a qualitative study of platform providers[29] Does the sharing economy increase inequality within the eighty percent?: findings from a qualitative study of platform providers[30] Does the sharing economy increase inequality within the eighty percent?: findings from a qualitative study of platform providers[31] Debating the Sharing Economy[32] Participation in the Sharing Economy[33] The Sharing Economy: The End of Employment and the Rise of Crowd-Based Capitalism (MIT Press): 9780262034579: Economics Books @ Amazon.com[34] Debating the Sharing Economy

I worked for about ten years in a survey research center that provided campus-wide service, though the client needed to at least pay what it cost to gather data for them. Projects from the University-affiliated hospital (College of Medicine, College of Nursing) could supply us with a *lot* of money.We started out as a 10? 12? station telephone survey facility that I cobbled together using computers even starving and broke graduate students refused to use. The chairs mainly came from a general-purpose, open-to-the-university-community computer lab I oversaw; since the computer lab was used for classes from time to time, extra chairs were needed for overflow, but we didn’t need them all.Then we really got loose with the money. We needed telephone modems in order for the computers to dial out at around 7.50 each (yes, 56,000 baud was/is still a real thing!). Next, we probably spent a couple hundred for tables and cubicle dividers from campus surplus (well, about 1/2 of the tables I actually re-purposed from a research facility that didn’t need them).So over the course of about two weeks, we created our first CATI facility. Humble, but functional! We did a lot of telephone surveys in there, but the size of the facility was frustrating, and some of the chairs actually emanated butt-smells on warm days. I replaced those chairs.Once we started to get attention around campus, we very quickly became a sub-unit of a University Institute that was well funded with grant money. They helped us acquire about 75% of a floor in a historic hotel the University has purchased. Renovations and customizations took at least six months (umm, do the wiring before plastering, then paint, and finally lay carpet; this being University laborers, the did things in exactly the reverse. They were slow, and needing to protect the carpet from the paint, repaint once the plastering was done, and so forth probably contributed to the slowness.When the project was finally completed, we were up to 35 stations using new computers, new chairs, new everything. Three of the stations were in a special room where they could be used to ask super-sensitive stuff like AIDS/HIV status.Anyway, your question. Obviously CATI (phone) surveys were important. We used Sawtooth WinCATI. It’s called WinCATI because they were so darn proud it actually ran under Windows, not just DOS. Even with the latest version (2014ish?), it still was kind of quirky and I needed basically to remember a bunch of undocumented trivia items to keep it running smoothly. We examined other options, and decided it was the least-bad package on the market. Perhaps most importantly, their support team was always accessible, always knowledgeable. So they would supply me with the trivia required.In the new facility, we actually had offices!We immediately began offering web surveys, mailback surveys, and for clients with deep pockets, CAPI surveys. We even carried out focus groups a few times. For such a small research center (four to six regular employees plus the contingent student workers for the call center phones) we provided a wide range of services.For the web, we used Qualtrics. One of the more expensive services, but powerful and flexible, with educational pricing not too bad.For mailback surveys, I would write web-based custom data-entry forms if I had time, or there was always Excel. We basically sucked at managing these. I wanted to get a good scanner or two and play with OMR packages to see what worked best. But, having blown so much on the facilities, the institute that oversaw us was surprisingly reluctant to give me $500-$1000 to do paper surveys right. In the end, we had a *very* dis-satisfied client who complained loudly enough that we outsourced mailbacks to a firm in California. They didn’t do OMR, they just had a facility filled with hard working Chicanas possibly being paid under the table. I have to admit, they were cost efficient and speedier than doing data entry with student workers. Probably more accurate too. Dunno how OMR would have compared.For one CAPI project, the client had two “programmers” available who wrote custom software so we didn’t need to purchase software. We used netbooks for capturing data. His “programmers” actually hindered operations, and there was a bad taste in everybody’s mouth when it wrapped up.For the second CAPI project, we used 4G-enabled iPad Minis. The data was entered into a very complex Qualtrics survey. Since we used address-based sampling, it was important to hit all the right houses and none of the wrong ones; also we needed to be able to track attempts and keep the Qualtrics data associated with the household especially in case of partial completions. So I wrote a web-based control panel. Surveyors could access the addresses they were supposed to hit, with a link to Google maps for a visual check. Then they would go to a page where they would record attempt information and such. If they got an eligible participant, then they could click on a link to Qualtrics; the Qualtrics ID was linked to the household ID in advance. Actually each household had two Qualtrics links,in case things went horribly wrong. The project manager could see everything the surveyors could, plus a little more info like live AAPOR response rates. Major problem: the cell carrier i had contracted with had horrible coverage. Happily, as a State entity, we could not be held to contracts, and we switched to a different carrier while keeping the iPads. Kinda unfair, but when did cell carriers play fair?Oh, for analysis, we mostly used Stata, plus a little SPSS. Sometimes we’d get fancy and check some measurement properties using Mplus.Hope this helps!OK, I’m back to describe how we functioned at my most recent survey research job. It lasted three years until I was eliminated as part of a company-wide layoff. After the job I described at length above, I was hired by a mediumish sized (approx 1,000 employees) non-profit. Although it was a non-profit, it definitely had a corporate culture created by all the MBAs in upper management working with management consultants. Which was fine with me. It required major changes in attitude, dress, work schedule, and my approach to problem-solving, but for what they were paying me compared to my pay at the University, these changes were more than acceptable.Our sole tool for administering surveys was Qualtrics, which became an issue for each annual budget: we needed to justify using such an expensive product instead of a cheaper one like SurveyMonkey. There are many reasons Qualtrics is superior, so we were able to continue using Qualtrics.A major difference from my previous job was that we made extensive use of Qualtrics’ reporting features. The reports Qualtrics could provide were generally too limited to be of use at the University (people wanted regression analyses, ANOVAs, and other models applied to their data, and nearly everything needed to have tests of statistical significance.) At my new job, univariate descriptive statistics such as frequency counts reported as raw numbers and percentages, averages, and perhaps some pie charts or histograms were generally al that was needed. Using Qualtrics reports saved us huge amounts of time and effort by automatically generating very attractive but very simple tables and charts that required little customization.Occasionally we needed to provide more complex statistics. The software used depended on who was assigned to that particular report. I would generally use Stata, but occasionally SAS; two of my co-workers were SPSS-only, and the boss was SAS-only.Apart from reporting, I used SAS extensively to generate samples for distributing surveys. I could have done this in Stata, but we were generally extracting the information from a huge collection of SAS libraries shared by about 100 employees. Also, as mentioned above, the boss was SAS-only, and he liked to understand exactly how the samples were selected. If the sample was not from the file in the SAS library (say it came from customer lists, or had been purchased from a third party) clients would generally provide Excel or .csv files I would make ready for use with Stata.Another use for SAS and Stata was ensuring that we complied with the CAN-SPAM act. At the University we didn’t worry about it, because we were often surveying faculty or students, professional organizations that supplied membership lists, or other lists to which CAN-SPAM did not apply. Besides, we were a governmental entity. At my new job, one could argue that our surveys did not need to comply since they were not commercial messaging, but some of the research we did for the Marketing department *might* be mistaken for commercial messaging, and the legal department wanted us to take the most cautious and conservative approach we could achieve. So we had several opt-out lists we applied to our samples, discarded email addresses we could identify as foreign because we did not know what regulations applied to what country (for example, Canada has a more stringent version of CAN-SPAM), and removed people who had been used in another sample during the past four months. While it may sound laborious, I kept SAS code and Stata code available that needed very little modification (generally just the name of the sample to be screened and the name of the desired output file). I felt that the Stata code seemed much faster to run and the command file was much shorter than the SAS version, but the SAS version was needed when working with extracts from the SAS libraries.We used Clarabridge for content analysis. This was seldom requested, but sometimes provided the most useful information one could extract from certain surveys with open-ended questions that motivated respondents to provide extensive answers. Realize that regardless of the software being used, content or sentiment analysis is a slow and painstaking process (unless you are applying a previously developed model to data that is gathered as a constant stream, such as hotel guests being offered the same survey upon every check-out at a sting of hotels world-wide). I found Clarabridge to be a little too automated, and it was clearly meant for use with data such as the hotel example described above. Extracting useful information from one-off surveys of the populations applicable to our industry required the us to modify numerous default settings and basically trick Clarabridge into doing things it would not normally do. Out of the software we used, this was my least favorite, and I wish we had been able to try other tools such as Atlas.ti.OK, thanks for reading this far; I never meant to write such a long answer!

Here is an answer I wrote back in May 2011. I believe the arguments still apply in the medium-term, although I certainly wouldn't say it's a great location to get long dollars here, and any next leg up in the dollar may not be led by a rally against the currencies currently the focus of the commentariat.Obviously the crude price has made a certain move since I wrote this piece, and may not continue to be a factor from here (also situation in emerging markets, not covered in this piece). Volumes are a different story, and the influence of the energy production boom on the US economy (at a macro level, and on the fortunes of energy consumers) remains far from exhausted.A couple of more recent pieces:Energy exports boosting US trade positionA jump in energy-related exports and a steep decline in oil imports lowered the U.S. trade deficit in December to nearly a three-year low.The improvement suggests the economy grew in the October-December quarter instead of shrinking as the government estimated last week.A brighter outlook for trade also illustrates how a boom in oil and gas production is reducing crude oil imports and making the U.S. a leader in the export of fuels. And it shows that higher domestic sales of fuel-efficient cars are lowering dependence on oil.The trade gap fell nearly 21 percent in December from November to $38.6 billion, the Commerce Department said Friday.Total exports rose 2.1 percent to $186 billion, driven in part by record exports of gasoline, diesel and other fuels.At the same time, imports declined 2.7 percent to $225 billion. That was largely because oil imports plunged to 223 million barrels — the fewest in 15 years.Read more: Domestic crude puts U.S. refineries in a sweet spot“While we may see some give back in the coming months … the trend of a narrowing petroleum trade gap will continue to drive improvement in the overall trade balance,” said Michael Dolega, an economist at TD Bank, said in a note to clients.Surge in fuel exports boosting US trade balanceHOUSTON — Growing production of U.S. oil and gas is helping to improve the nation’s trade balance, according to a federal report Monday.Dramatic growth in the export of refined petroleum products, such as jet fuel and gasoline, has led the way. The value of net refined exports increased 55 percent in 2013 over the prior year, reaching $33 billion, according to the U.S. Energy Information Administration.U.S. refiners are finding cheaper domestic alternatives to overseas oil, causing a rally in the ratio of refined fuel exports to imports. Overall energy export values increased 8 percent in 2013 over the prior year.Total energy imports to the U.S. fell by 11 percent for the same time period.Trade balance: Obama falling short on exports, but fuel shipments boomingThe shifts have helped push down the U.S. trade deficit to its lowest level in four years, because of the importance of energy imports and exports. Energy accounts for 15 percent of gross goods imports and 7 percent of gross goods exports, the EIA said.The increase in production also has dampened imports. Natural gas imports fell 14 percent in 2013 over the prior year and net crude imports fell by 16 percent.Going forward, one big driver that has not been of importance previously in this recovery and dollar rally will be the labour market - in particular wage growth - and the impact on interest rate differentials. Stronger US growth means rising yields, which on balance will tend to be dollar supportive.It goes without saying that this is not intended to be investment advice.I recognize that the length of this answer goes beyond being of merely unfashionable length - it was written for a different audience originally - but perhaps it may be of interest to those who have the interest and time to think about macro trends (like the outlook for the dollar). One ought to make things as simple as possible, and not simpler, and unfortunately currency valuations are not quite the simplest phenomenon in the financial world. It may also serve as some use to illustrate one less common way of thinking about markets that did pay off in this particular instance.=============Since the Federal Reserve trade-weighted nominal broad dollar index peaked at129.68 in February 2002, it has depreciated by 27% to its recent monthly closing lowin July of 94.84. Although inflation since that time has been significant, it has nottended to be lower in America’s trading partners, and the depreciation in the realbroad exchange rate over that time has not been significantly different at 28.6%.There is overwhelming agreement amongst financial commentators that this declineis not only fundamentally justified, but that it is only the beginning of a much largerdecline that presages the end of America as the dominant empire, the destruction ofall nominal values amidst money-printing induced hyperinflation and a breakdownin and disintegration of the social and political fabric that constitutes the nation.There is simply no prospect of a stabilization in the triple deficits faced by the US –negative private savings, a worsening fiscal deficit and therefore a continuallydeteriorating current account. The only option is to save oneself by investing inforeign currencies, foreign and equity markets, commodities, precious metals and inmore exotic hard assets such as art and infrastructure.Not only have many foreign nations (emerging markets in particular) improvedtheir infrastructure and quality of their governance, but they are now clearlychallenging the US for its former position of supremacy, and to listen to someauthors there is not even any longer a challenge – the US has lost its place, and notonly cannot regain it but does not deserve to. As investors have lost confidence inevery institution and strategy that they once held sacred, they have abandoned whatis abstract and looked to what is concrete – when one visits the emerging world onecan see for oneself the very radical and genuine changes that have taken place andpersuade oneself that an even brighter future lies ahead.I would like to consider the merits of this view more carefully, both from a shorter-term investment horizon of one to two years, and from a longer term one of a decadeplus. I will start by considering the question of valuation of the US dollar vs othercurrencies from a couple of different perspectives; it is true that in extremis one canwait for a lifetime before a currency returns to fair value, but I have always found ita useful starting point to ground the analysis. There are certainly more sophisticatedmodels than purchasing power parity (PPP) that incorporate other importantinfluences, but I shall focus on this simple approach to valuation for the time being.[PPP table of dollar valuation based on 3 different metrics - 1 shown here]This suggests an undervaluation of the US dollar by approximately 35%. It ispossible that the Big Mac index is biased by relative factor availability (cheap butrelatively unproductive labour is abundant in poorer, less developed countries, andthis tends to bias the results to make their currencies seem cheaper on PPP-basedmeasures of valuation), but I note that this gives a result close to that obtained if weuse the OECD valuation measure on its own (which suggests an undervaluation of30%). Interestingly this suggests that the dollar was actually a touch cheap in 2002 atthe beginning of the latest bear swing.There is a striking difference between the results of this analysis and the popularconception that the dollar needs to depreciate until US wages reach Chinese levels.This conception simply is not founded in reality since it does not take into accountrelative productivity differences. Proprietary analysis by the ISI Group shows theUS to have very competitive unit labour costs internationally - lower than Germany,and comparable with South Korea. Anecdotally we see many signs of the re-localisation of production that tend to confirm this thesis - Indian call centres havemoved activity back to the US, and the difficulties of doing business in Chinacombined with the high rates of wage inflation mean that at the margin a few yearsforward the US becomes a more attractive location. In essence the first wave of thepost-internet global labour arbitrage is complete, at least as far as the US isconcerned.Beyond exploring valuation, I have over the years found it useful to consider the roleof sentiment in evaluating the attractiveness of an asset. In investing we are tryingto buy cheap assets. Since any deviation of an asset can be considered a 'riskpremium' of some sort, we are trying to identify assets where the risk premiumreflects an irrational perception of the risk or of impairment to the underlying valueof the asset. This means that we prefer to buy an asset that we can identify as beinghated, and we prefer to sell assets that are universally loved. Usually a hated asset ishated for good reasons, but these may be in the price or it may be the case that thingsare about to look up and that the underlying factors weighing on the price weretemporary.How would one assess sentiment towards the dollar and towards the USA (peoples'feelings towards currencies often reflect how they feel about the issuing nation, sothere are always aspects of national prestige mixed in with the hard, coldcalculations of modern finance)? I would say that it is unbelievably grim. It wouldbe hard to imagine sentiment being much worse than it already is. Here are somerecent cover stories from the Economist magazine. Note the themes of fighting anddisintegration (manifestations of a low social mood that tend to be found closer tomajor lows than to major highs)But perhaps, despite its undervaluation, there are good reasons to be diversified outof the dollar into other currencies. Here are some of those that come to mind, andhave often been raised by the proponents of permanent dollar weakness.•Chronic current account deficit•Unmanageable debt load (public and private) vs GDP•Japanese experience suggests deleveraging will require monetary policy supportfor years to come to support GDP growth. QE 3, 4, 5 ?•Very polarized political scene, disturbing lack of social cohesion suggests takingtough decisions will be hard. Majority of households do not pay income tax! Politicalswing away from policy elite towards a more populist direction (‘Tea Party’ascendancy). Republicans playing chicken with Democrats over the budget – risk ofthreatening US credibility and prestige in bond market. Ongoing black swanpossibility of a technical default?•Secular decline in US civilization accompanied by a genuine ascendancy of theemerging world. Declining moral qualities (restraint, willingness to sacrifice for thefuture), loss of optimism. Is the American Dream dead? Wages in emerging worldstill cheap in USD terms vs those of US workers.•Stagnant real wages since 1970s; declining overall participation rate since 2000 –very much worse for men, particularly those from minority groups (that are in thesample statistically less skilled).•Iraq, Afghanistan, Libya. Wherever next ?Mass perceptions of national prestige are not wholly irrational, but they are oftenshaped by images. The past decade has seen a striking contrast in the developmentof major US cities when compared to what has taken place elsewhere. Some havetaken these as signs of a secular decline in US civilization.Compare the Lee Plaza Hotel in Detroit -to the still very-rapidly developing Shanghai skyline:The tallest US building was completed as far back as 1973 (unlike in the Arab worldand in Asia where new highs are set quite frequently). Possibly this reflects a loss ofambition and optimism in the US mind.The Austrian economist Bohm-Bawerk, in his Positive Theory of Capital, observedthat as a civilization develops its members become more and more able to considerthe longer-term consequences of their actions today for the future, and to restrain thegratification of their present impulses in order to have greater fulfilment at somefuture time. He linked therefore the development of a society with rising future timeorientation and a rising personal savings rate.Moral Decline - End of Self Restraint?The decline of the personal savings rate as we see in the chart below hascertainly accompanied changes in culture and personal conduct that could beconsidered jointly a consequence of greater contemporary American emphasis oninstant gratification. If I am right, then a consequence of this has been also theneglect in the political domain of problems which will eventually become very largebut have not yet bitten.Declining future time orientation reflected in personal savings rateThe current account by the National Income equations is identically equal to the sumof household savings, corporate savings and government savings. It seems strangeto speak of government savings in our era, when the deficit/gdp path seems to havebeen on a one way trip southHorrible Deficit/GDP PathAs we would expect, this has been reflected in a deteriorating trade balance andcurrent account.Declining Trade Balance/Current AccountAmidst much talk of the hollowing-out of US manufacturing and 'jobs movingabroad', we have heard talk of the declining labour force participation rate:Declining Participation Rateand the picture is even worse for men (it is particularly bad for some minoritygroups and for lower-skilled workers)Picture worse for Men (particularly bad for some minority groups and for lower-skilled workers)Now at least we have a good grasp of some of the very negative longer-termfundamentals for the dollar (although I have not addressed questions of debtsustainability at this juncture). Having given this concession to the dollar bears, letus consider whether these bearish arguments are in fact complete.Reality Check•Is the current account situation really so bad? We ought to examine the detail.•Bears on the US want to have it both ways – to believe that prospective US growthwill be weak due to dampened animal spirits, and that the current account willcontinue to worsen leading to further dollar weakness.•But dampened animal spirits ought to lead to rising/high savings rate and astabilizing/improving current account. Not likely to be associated with largeportfolio outflows on the capital account either.•Danger of extrapolating recent experience indefinitely out into the future. Everynatural phenomenon has an ebb and flow. Possible we might see a stabilization orturnaround in participation rate, government spending/fiscal deficit and nationalself-confidence.Please look carefully at the charts that follow. Further discussion below.Recent evolution of savings rateTrade balance ex Petroleum (bn)US Trade Balance ex Petroleum (%GDP $)US Trade Balance Petroleum Products Component (Current $)Petroleum Balance (estimated volume)US Crude Imports (Value)US Crude Imports (Volume Proxy)So despite the gloomy trend for the evolution of the savings rate, looking at the morerecent data indicates a potential change in trend. Given the horrible scare thatpeople have experienced, it is likely to be quite some time indeed before they feelready to live once again on the never never and they are likely to want to rebuildtheir balance sheets by accumulating savings - so this is one positive aspect of thebrutal period we have lately experienced.I broke down the trade balance into two components - petroleum and ex petroleum.If we look at the ex-petroleum part, we can see that this has actually been improvingeven in nominal USD terms for quite some years - what has led to the deteriorationof the overall number has been a steady deterioration in the component coming frompetroleum products.Drilling down once again, we see that this is a consequence not of risingconsumption, but of a rising oil price. (As every schoolboy knows, crude has been ina bull market for the past ten years). So, if we want to understand better theprospects for the deficit, we need to consider what is likely to happen to US importsof crude and to the price of oil.Until recently, any discussion of the outlook for the future availability of crude oilhas been dominated by the perception that we have reached peak oil, and that fromhere the output would inevitably decline, implying a bleak future of tremendouslyexpensive oil (given the very rapidly-rising demand from the emerging world) withsubsequent rationing and resource wars. Under so-called Olduvai theory, the wholeof industrial civilization was a mere blip between two periods of primitive barbarianliving conditions.W.D. Gann spoke of an economic and financial cycle that involved old men passing away and new men coming to power. I have observed previously how the timing of the passing away of important figures tends, for whatever strange reason, tosynchronicitously reflect major shifts in society and the economy. From thisperspective, it was intriguing to see that Matt Simmons, the most respectableproponent of Peak Oil Theory passed away last year.What shifts might there be with regards to social perception of the outlook forenergy supply? One striking development over the past five years has been the newapplication of a process for extracting natural gas from shale. The surge insupply (amidst weak demand conditions) has been associated with a collapse in thespot price of natural gas that one could barely have imagined just a few years back.We are now seeing increased output of shale oil. Last year saw an upturn in UScrude output for the first time since 1986.Shale Oil (Tight Oil)Natural Gas and Crude Oil ProductionNatural gas production at all-time highs. Upturn in US crude output for first timesince 1986.Collapse in Natural Gas Prices following Shale Gas coming on lineCombination of rising supply (shale gas), limited storage availability and still-depressed US electricity demand vs peakUS Shale Output (Liquids k b/d)Per a US consultant, US liquids output could add 700k b/d in Gulf of Mexico, 200kb/d in biofuels, 100k b/d oil from natural gas formations, 400k b/d natural gas liquids. So they estimate a total increase in US liquids output of 3.3mm b/d by 2015.How significant is 3.3mm b/d increase in US production?Net Imports–1973 6.1 mm b/d–1980 6.4 mm b/d–1990 7.1 mm b/d–2000 10.5 mm b/d–2010 9.5 mm b/dSo very simplisitically, this order of magnitude increase would reduce the net import requirement for crude by 1/3. The gross trade balance was -$48.2bn; ex petroleum it was -$16.9 bn. Presuming no change in price and everything else held the same, this would reduce total trade balance to -$27.3 bn.But supposing we were to see a correction in the price of crude (something to be expected given increased supply, but I have explored elsewhere some reasons relating to monetary policy as to why this kind of fall might be possible) by as much as 40%, it would put total trade balance at -$23bn – levels of 1999 in current USD, and even better as a share of GDP.In 1999 DXY was 100, 32% richer than current levels.So here there is a scenario that could lead to a substantial improvement in theperception of dollar fundamentals. One needs to have a degree of imaginationrather than awaiting all of these events to actually play out since otherwise the movewill have already taken place by the time one is sure. Market timing and riskmanagement become key in profiting from such developments.One should note also that for now there is a strong relationship between equities andthe dollar index.Equities (SPX) vs even-weighted dollar indexFor now still more driven by risk than interest rate spreads. Correlations do change over time.If we look at the relationship between EUR/USD and interest rate differentials, we can see that it is a bit tighter.EUR/USD vs US/Germany 2y SpreadEUR/USD clearly influenced by interest rate differentials. Rally in Euro since last year was associated with sell-off in schatz and inflation fears.So if I am right in my article elsewhere in the idea that the Eurozone can catch up to the slowdown in global growth, interest rate differentials could turn very much more supportive for the dollar.Outlook for EUR/USD•Depends on risk appetite overall•Scope for pricing out rate hikes as periphery continues to deteriorate, particularly ifinflation concerns recede.•Greece cannot possibly pay back its government debt.•Nascent shift to radical right means that time is running out for EU integrationists.Bailouts politically unpopular, and EU fiscal union seems very far away.Last time crude prominent in news (May/Jun 2008)How did it turn out last time? (Crude)A little speculative froth?Last time sentiment towards USA this bad? (Loss of prestige)"America is the Great Satan"Dollar IndexUS Consumer Confidence - still very weakDespite improving labour market and a strong equity rally US consumer confidence remains remarkably weak. Historically this has been a good contrarian indicator (buy stocks when confidence hits depressed levels).Vicious Cycle Turns Virtuous?Why is Consumer Confidence weak? Perhaps because of elevated gasoline (and other commodity) prices. This has been a focus of much popular discontent towards the Fed.Most bullish phase for commodities is abundant liquidity and relative weak growth. Ie crude is high because dollar is weak and because the effective Fed Funds rate is low. The effective Fed Funds rate is low in part because consumer (and business) confidence is weak. Headline and core inflation both currently low. Tightening in rental market (vacancy rate was 12%; now 9.5%, 2012 5%) implies pickup in the large OER and tenant’s rent component of CPI. Core inflation 1.8% by year end 2011 according to DB.QE2 ends shortly. A pickup in hiring and spending would be associated with upward revisions to interest rate expectations, which would tend to support the dollar and weaken gas prices – which would tend to lead to improving consumer confidence.Postscript (June 2015)We did see a dollar rally and a sell-off in crude, and really the question is what happens from here. Angst about Europe is certainly in the price, and I wouldn't suggest that it is appealing to focus bearish views on this region (one has to be bearish another currency to be bullish the dollar). Similarly, tight oil is now on most peoples' radar screens, although I wouldn't suggest we have seen the full extent of bullish-US economy implications from this development (even though other nations have also discovered tight oil that is potentially commercially exploitable).I think from here the points about the relative international competitiveness of the US remain important. It simply isn't the case that US workers cannot compete with Chinese workers until we are all earning 50 cents an hour because for certain reasons, mysterious and well-understood, US workers remain very much more competitive than many other workers (and in the meantime China is seeing very significant wage inflation).Unemployment continues to fall, and soon enough (I happen to think rather soon indeed) we will start to see the implications for nominal wage growth. This will tend to be bearish US fixed income, which will lend interest rate support to the dollar. (Bonds will sell off not on news of the first hike, but well in anticipation of it. The Fed always follows the market).The US has been called the Saudi Arabia of agriculture - it's a very competitive producer in this domain. For the time being this hasn't made much of a difference in recent years at a macro level because prices are not much more than 5% of the levels of the 1970s peak in real terms. Agricultural prices do experience profound swings over long periods of time. If I am right, we have a monster bull market coming, and this will be very positive not just for US farmers, but for the US itself. The reasons why are varied - Jim Rogers has set out some important reasons, but if I am right that we are early in a cooling period (I eagerly await your downvotes), then there will be implications over time for agricultural prices. The Year without a Summer did occur during the Dalton Minimum, and wasn't an easy time for agricultural production. Conditions are different now, but is it truly the case that the output of the sun has no influence on crop yields in the modern era?It does seem to me also that the US is beginning to regain its nerve. That's really key, because without confidence entrepreneurs do not take the risk of starting new enterprises, and they do not hire people. Similarly, without courage in the political domain there can be no hope of addressing problems at their root - in many cases the origins of these things goes back many years. Often it is only when one truly has no choice that one is forced to confront problems and work on solving them. That is the lesson of the peripheral European countries that have amidst considerable hardship begun to take painful decisions to fix things, just as Germany did in 2002 when she was considered the sick man of Europe.The dollar is overbought here - I wouldn't buy it for a trade. But I doubt it will collapse against other currencies over the next few years.Btw in case not clear, I certainly would not be long the broad usd here (nor the DXY, which is heavily eur weighted and the eur is hated for bad reasons), I think these arguments above hold in the medium term, particularly in as much as I don't think the dollar will crash.But in a long term bull market for a complex you see tremendous rotation. Sugar had two major bull markets in the 70s yet gold didn't peak for many years later, long after sugar did. And it may be the same with the USD.Possibly Asian currencies are next, although I have not been following markets closely. The global excess savings phenomenon and the conundrum was really a US excess liquidity phenomenon with the U.S. current account, FDI, and long term portfolio outflows being the counterpart of Asian reserve acquisition of UST. Given the impact of energy (today) and perhaps higher agricultural prices (in coming years) on the current account and also renewed private and public conservatism on national savings rates, and the impact of rising yields on the attractiveness of the dollar, we may see appreciation pressure on usd vs Asia ex japan. This will put pressure on reserves, and liquidation may support a trend of rising yields, which itself will tend to support the dollar.This being said, the relative yield differential theme may be more supportive even for the EUR than the USD, so it becomes less of a pure USD trade at this point.I will write more on this shortly, probably not on Quora.Of course in purchasing power terms it is a different matter (see my answer on this). This certainly is one domain where it doesn't pay to bet against the Fed. Rather than price stability, the Fed has obliterated the real value of the dollar against the cost of living in the past century odd since it was founded. I am not sure things are going to be very different in future.Update March 2016: I am not bullish the dollar any more.