f-droid doesn't have payments, Amazon is yet another faceless mega-corp that has had its fair share of sellers accusing them of various shady things.
Creating a Play Store alternative is hard for many reasons:
1) finance: the more countries you operate in, the messier it becomes. International taxes are one hard mess, KYC/AML regulations differ between countries, and to top it off there is the whole "international sanctions" issue especially regarding Iran (where it's fine and explicitly encouraged by the EU to do business with Iran, but any entity that has US exposure exposes themselves to liability in the US for violations).
2) vetting of apps against a constant onslaught of spam, malware, copyright violations: f-droid has it a bit easier since they require all apps be open source, but a commercial, widely used alternative will have to run static analysis, dynamic analysis (to catch runtime exploit attempts) and manual testing. All of this is expensive and requires expert knowledge of Android as well as IT security.
3) Implementation and hosting: an app store worth its name has a lot of binary assets to distribute to users (and again, you have to avoid getting into trouble with people abusing your service to spread illegal content, because there will be such cases rather sooner than later), the store itself has to be implemented, regularly adapted to account for changes in the Android core, you definitely want a focus on security to avoid some hacker distributing malware to all your users with a push...
4) Customer and developer support: it's a well-known meme that FB/Amazon/Twitter/Google are almost impossible to reach for ordinary people without raising a shitstorm on Hacker News or a well-funded lawyer team... but the key thing is, support is expensive to run.
Developers have no incentive to go the extra mile to publish on your store, because there are no users and it is extra work.
Users have no incentive to install your store, because there is barely anything on it.
OEMs have no incentive to preinstall your store, because you don't have any content (which devalues their product) and they don't gain anything from it. If they are willing to roll their own store, they at least rake in all the profit from it.
Developers like in-app payment methods because all they have to do is integrate an SDK for payment, and then they get a monthly payment on their bank account and a bill for accounting, that's it.
If you want to handle payments yourself, you'll have to:
1) implement user management to deal with storing what stuff a user has purchased, with all the GDPR and customer support (forgotten passwords, hacked accounts, lost MFA creds) headache that comes from that
2) find a payment processor that operates in all your target countries (no, Stripe and Paypal alone won't cut it), and integrate these (and hope they don't run into the same issue with Paypal, who are known for deciding on a whim to withhold funds)
3) Issue individual bills to customers, account for stuff like cross-border VAT, insanities like county/city sales taxes, deal with refund laws
4) deal with fraud attempts, angry parents, ...
5) Re-implement recurring payment schemes if your business model wants these
In the end, app stores (and ad SDKs) are a matter of convenience. Big shops like Epic, Spotify, Netflix can get away with running lots of this infrastructure on their own, but 99% of small devs don't have the time, knowledge and legal requirements to deal with that.
Creating a Play Store alternative is hard for many reasons:
1) finance: the more countries you operate in, the messier it becomes. International taxes are one hard mess, KYC/AML regulations differ between countries, and to top it off there is the whole "international sanctions" issue especially regarding Iran (where it's fine and explicitly encouraged by the EU to do business with Iran, but any entity that has US exposure exposes themselves to liability in the US for violations).
2) vetting of apps against a constant onslaught of spam, malware, copyright violations: f-droid has it a bit easier since they require all apps be open source, but a commercial, widely used alternative will have to run static analysis, dynamic analysis (to catch runtime exploit attempts) and manual testing. All of this is expensive and requires expert knowledge of Android as well as IT security.
3) Implementation and hosting: an app store worth its name has a lot of binary assets to distribute to users (and again, you have to avoid getting into trouble with people abusing your service to spread illegal content, because there will be such cases rather sooner than later), the store itself has to be implemented, regularly adapted to account for changes in the Android core, you definitely want a focus on security to avoid some hacker distributing malware to all your users with a push...
4) Customer and developer support: it's a well-known meme that FB/Amazon/Twitter/Google are almost impossible to reach for ordinary people without raising a shitstorm on Hacker News or a well-funded lawyer team... but the key thing is, support is expensive to run.