It seems to me that the question being posed is actually: why do employers prefer educated workers, even if the education is not immediately related to the job?
I'd advance the following possibilities:
- General skills conveyed by education;
- Measure of personality;
- Socioeconomic background;
- Mismatch between educational fields and occupational divisions of labour.
General skills conveyed by education
One traditional distinguishing feature of formal education, as distinct from any trade training, has been that the conceptual content of a field of study is discussed and analysed explicitly, verbally and in writing.
These activities involve cognitive skills relating to verbal and written communication, and the organisation and critical scrutiny of information and concepts.
Stereotypically, the difference between the educated worker and the trade worker, is that the educated worker can tell you how things are done (and why they are done, and why things are not done otherwise), whereas the trade worker can only show you how things are done (and perhaps, not done).
I'm not myself implying any superiority in the understanding possessed by the worker in one mode or the other, but for those concerned with management rather than personal execution, the ability to formulate then discuss plans (submit them to collective examination) and express commands (and convincing reasoning) to coordinate a workforce are also important.
For those participating in occupations where there are constant complicated variations in execution (and thus a great deal of management labour required), they might not need to make and articulate complicated plans themselves, but they need at least to be able to hear and understand those who do (and scrutinise the content enough that the managers are not allowed to run amok with corporate resources and incoherent plans).
For many employers, they need a workforce in possession of the cognitive skills described above. These skills are far more important than the detail of any particular field of study pursued in education, and whilst there are wide margins of error, those with an education will on average have better skills in these areas than those without.
Measure of personality
The pursuit of an education has traditionally correlated with an intellectual or studious personality.
Or at least, if not "personality" in the true sense, then certainly habituation and trained preference for certain kinds of activity.
Employers recruiting for desk jobs want people who are happy to be sat down most of the day, thinking, writing, talking (about work), or concentrating on some kind of detailed activity.
Often we're not necessarily talking about "intelligence" here. Modern corporations often have a need for those who do the clerical equivalent of stacking shelves and digging holes with shovels, but without the physicality.
A significant number of those who perform modestly in education, or who perform well only with the input of undue time and effort, will be sought out by employers to fill these kinds of sedentary routine jobs.
Socioeconomic background
In Western society there is still a correlation, but especially in underdeveloped societies like India, I suspect the possession of a university education has a considerable correlation with socioeconomic background.
This is relevant to employers because it implies the worker's parents will have had the resources (money, energy, free time) to raise the worker to a good standard of culture, and that the worker is more likely to have the conservative attitudes resulting from an upbringing with relative stability and with an absence of the most intensive economic exploitation of the parents.
Because capitalist relations are defined by exploitation, this is always a relevant consideration, but in sectors defined by a high degree of industrial conflict, the employer may be explicitly recruiting for those least accustomed to thinking which challenges the structure of the existing system and least prone to collective organising or a class mentality.
Mismatch between educational fields and occupational divisions of labour
Often, curriculums in education do not map directly to the needs of a particular employer filling a particular role.
This is because workers are reluctant to embark on multi-year studies that tie them very closely to only a particular economic role, and because employers, never offering jobs prior to the completion of study, have little ability to express demand for study to match their exact needs.
Even where an employer could expect to recruit someone with a generally relevant field of study, they probably find from experience that they are not so over-supplied with candidates that they can always get one on the spot - and they'd prefer to have bums on seats and work moving slowly, than not at all.
Because fields of study are defined with somewhat arbitrary boundaries, and because divisions of labour are often not such as to require only specialist knowledge and nothing else, many employers also find that they can start to suffer from monocultures if they recruit only educated workers who have studied particular fields.
This is where some kinds of specialist knowledge are over-supplied in their workplace, and other kinds of knowledge are under-supplied, or where there are few integrations of knowledge across different fields which might be extremely valuable to an employer.
Essentially, the need of educators to divide up fields of study for the purpose of teaching or study (especially over a limited timeframe), cannot be expected to align perfectly with an industrial division of labour.
Conclusion
The field of study is often of least concern to employers of educated workers.
What matters more is:
The skills conveyed by the process of education generally;
How education implicitly selects for certain features of ability or mentality which employers find desirable in workers; and
A certain amount of intellectual diversity in the workplace which undoes the harmful (but necessary) specialisations which formal education imposes.