You're correct that manufacturing too has a finite demand but it still has an order of magnitude more potential than the service industry. The vast majority of service jobs are extremely local or tied to a language/region/jurisdiction so those jobs have very little to export. Meanwhile, the vast majority of the world is not yet affluent enough to afford most luxury manufactured goods so as the population gets wealthier, the market for manufactured goods expands. As long as the developing world follows western consumerism, and by all vounts it looks like it is, there is a lot of room to grow. Developed countries who have centuries of industrialization behind them and a knowledge economy stand a much greater chance of capturing that newly generated wealth because of scale and experience. As long as higher labor and regulatory costs can be offset by automation, the manufacturing can be grown with the world while services can only really grow with the national population.

I see a lot of growth potential for manufactured goods if you measure them in mass. I'm not sure there's great growth potential if you measure them in money. Every wave of industrial relocation so far has overtopped the previous one by lowering costs. China's steelmakers are exporting more tonnes of steel than US steelmakers did 50 years ago, but they're also commanding significantly less money per tonne. For that matter, if poor countries are going to get wealthier but buy manufactured goods from abroad, where is their new wealth coming from if not manufacturing or services?

You have an over-narrow concept of 'services'. A large part of 'service' wealth creation happens in management of supply chains (import/export, sales, marketing, logistics, right down to shelf stacking and warehouse workers) which scale alongside industrial demand, not population. Overseas manufacturers can't access markets at all without effectively paying service providers to help them distribute their goods. Walmart is, in a way, an exporter of market access services.

That's even before you consider service industries that create intellectual property which scale with global laziness and boredom, or the finance industry which provides services that scale with global economic activity, rather than local population.

Perhaps you wouldn't, but I do know people who regularly change out all their furniture and appliances for newer models that have more bells and whistles and/or are more "fashionable".

Would they do it every single year if the furniture and appliances got cheap enough? IMO, yes.

I would argue this is done more as a psychological status symbol. If goods such as furniture and appliances are so cheap that your "average joe" can buy the latest appliance on a whim, the symbolism becomes diluted and your friends will be less inclined to make frequent purchases for these items. They will seek out new status symbols.

While the number of goods might be limited, the quality is not. If you have more disposable income, you are just as likely to spend it on a better car, a better house, a better laptop, a better phone, etc. Higher quality goods require more complicated manufacturing, more resources, more workers, and longer supply chains. All of which employ more people.

I'm not sure the quantity of goods is limited either. Looking at a supermarket today, the diversity of goods for so many niche purposes is incredible. And the internet has increased it an order of magnitude more.

These high quality goods require more sophisticated automation, not more workers. Unless you expect a future of hand made items.