I think allocating money properly would be creating value. Deciding were to invest could be one of the most important decisions possible. That many finance types might have been involved in something else is another matter.
Even if you want to manufacture stuff, you have to somehow organize the manufacturing. To say only manufacturing is worthwhile seems a bit shortsighted. Suppose all the finance guys would go to GM and work in their factories assembling cars, would it help much?
Allocating money properly is not creating any value at all. It's just a regulatory service that works according to some rules - it's a bürocratic step in economic system, and though it is possible to create some valuable people by then developing skills or experience, but they don't create value. They move value around.
Allocating money (or time or work hours) is a necessary step in creating anything. I don't think the hard nosed approach to estimating value is very useful. You could say that only the actual workers producing something physical create value, but at the end of the day, all people in the chain contributed to it.
Consider a scenario where you have two alternatives: you could ship shit from one pile to another for 5 hours, or you could spend 5 hours brewing beer. Suppose the worker has no idea whatsoever as to which activity is more benefitial, so the finance guy decides whether 5 hours will be spent on moving shit around, or brewing beer. Suppose the value of beer is much greater than moving shit - I think at the end of the day the finance guy will have contributed to bringing the beer into existence.
I know that there is this school of economics that tries to define value by the amount of (physical) work that went into it, but it really doesn't make much sense. You don't create value simply by doing something physical - it has to be useful, too. So I think the usefulness of the end result is a much better estimator of value.
Okay, it seems that what I'm trying to explain is difficult for you to understand. Let me change this a bit.
Imagine you are made president of a 3rd world country which has not yet industrialized. You are now tasked with making this country actually rich. Where would you put the government money in?
If you say you would equally create service and manufacturing industries, then your country would collapse quickly. The correct thing to do is to create goods that can be exported, or that have some use within the country itself. The other industries always develop themselves, because they are a sort of grease for the core industries.
Allocation of time and money is not done by bankers. It's done by managers and owners, the bankers just approve or disapprove loans.
I think bankers have to decide if a loan is promising or not. If you want a loan to start a business that is doomed, they should say no. Without money, the managers can't do anything.
I don't want to defend all bankers, but I think it could be a service like other services.
Your 3rd world example also falls short. I wouldn't create a "planned" economy, but go for demand. Actually a lot of 3rd world countries fare reasonably well with tourism, which is maybe more service than production? India also exports services - it is possible, so I think the generalization "products over services" fails. Both are necessary.
If you discard the important finance decisions to do with not only the allocation of resources, but also protecting your country from outside financial 'wars' then you could be creating so much value but not able to retain that value, just look what happened during the Asian financial crisis.
A free-market economy without banks will have an awfully hard time starting new enterprises; pooling savings and investing in businesses is the purpose of a bank. Managers and owners can't allocate time and money without capital.
Even if you want to manufacture stuff, you have to somehow organize the manufacturing. To say only manufacturing is worthwhile seems a bit shortsighted. Suppose all the finance guys would go to GM and work in their factories assembling cars, would it help much?