Productivity is a time-based concept. Whether your pricing unit is a word, a page, a line or a character, it takes a certain time to produce. And there are limits, particularly ceilings.
An interpreter outputs 150-180 words a minute. Processing is probably faster -- a concept flies off to its equivalent concept in another language through the mind almost instantaneously -- but the interpreter has to discipline his tongue so as not to become unintelligible while s/he delivers the message at a controlled pace, sheathed in solid grammar. I am sure everybody can think of some speaker (Dustin Hoffman comes to mind) who they are convinced should be receiving speeding tickets, but that person’s counterpart in the X language cannot be racing him to the death, concentrated as s/he is on making the content of his discourse understandable (which is not the same as word-for-word rendition).
When a person sits down to a keyboard with similar intent, s/he also deals with upper limits of a similarly mechanical sort. Perhaps s/he can afford to pay more attention to the individual word (while watching out for overtranslation), but there will be an upper limit to what s/he can do in any given unit of time.
When I began in the business, an optimum speed for translators working in conferences was set, to be both safe and reasonably demanding, at 500 words per hour. (Note that translators were screened with a view to building ad-hoc teams for conference production lines, not necessarily a process that agencies engage in). Revisers, on the other hand, were expected to output between 1200 and 1600 words per hour. They were not expected to dally -- the translators had been screened, after all -- and four interventions in that working hour were considered reasonable. Translators and revisers received the same per-day compensation, with a view to eliminating any conflict-of-interest factor.
Tools have been developed since that would tend to alter the separate components of these ratios. While posing something of a hurdle for fast typists working on non-repetitive texts, translation memories, in some cases, have enabled considerable improvement in terms of daily output, boosting the 2,500 or 3,000 words a day that can be achieved without them. And, whereas voice recognition is still in a relative minority as a tool for translation, it nonetheless holds a certain promise for those willing to invest in its development. (Buying the outlay is easy -- I am referring once more to an investment in terms of time, as one "trains his Dragon" or equivalent software).
Hence, even faced with the principle of uncertainty, it becomes possible for freelancers to set targets: to determine how much they want to earn and pay in taxes, how many days a year they want to work, and how much time a day they want to devote to themselves and their families.
How much is your time worth?
That said -- and the product having been considered from the point of view of time -- it now remains to be seen what one’s time is worth. Nature has been fair enough to endow each and every one of us with 24 hours a day, which might be inferred as divine, heavenly, or sidereal justice, depending on the view we take. What remains clear, however, is that we sell a limited resource that makes our product more akin to fossil fuels from the producer’s point of view than it is to die-cut components justifying volume discounts.
An hour is an hour, whether it is spent in New York, New Delhi, or Kiribati. Globalisation would hold that places should reflect specific differences, but this has its limitations in language services. In the first place, the self-same information technology that makes such an assertion possible does not seem to have place-specific discounts: computers sell everywhere at roughly the same price, with the occasional bulk discount made precisely to those distributors selling in areas of heavy traffic, where people can afford it. Hence, where subsistence wages make the capital outlay in hardware prohibitive, we will not be likely to find a healthily thriving community of translators competing with each other to dump prices -- not over the long term. Because, if we follow a simple logic that a given country with a lower wage standard and cost of living should sell at lower prices, said given country eventually hits the point where planting potatoes and internal trade turn out more profitable than maintaining an effort to sell quality language services to the world at large.
In the second place, and despite many official policies tending to transform certain parts of the world into fortresses, the industry does set certain proficiency requirements that make travel and the attendant expenses practically essential, entailing periodic investment in displacement and language maintenance / language updates. This part of the package is not to be had for the local price of potatoes, and despite arguments disclaiming the need for travel, or the need to pay for such a component inherent in the service, it is in general a happily accepted source of relief when, without mentioning that he pays for it, a client is assured that, yes, his translator has indeed travelled to, or lived in, the country or countries where his source or second language is spoken.
In the third place, time has a certain cumulative added value. We may argue this point with EN 15038, but five years of full-time freelancing are not to be sneezed at until one has tried to get there. There is a valid reason behind the EU asking its applicants for their tax declarations for the past five years in lieu of a degree. The economic survival factor is not to be sneezed at, since no one flourishes long in the field with a bad product, and bottom lines may well determine motivational leanings. It is also often said that translators and interpreters age like wine, principally because experience is also cumulative.
Hence, it should come as no surprise if older or more experienced translators say "no" to arrangements that can do them more harm than good, that do not comply with their ideas of working conditions, or that are simply not cost-effective. And perhaps rates would be their most efficient recourse for stipulating those conditions that they want, by way of risk filters and defensive pricing.
Spain