FDA approval is slow and expensive, but the political pressure to change that situation has created myriad alternative routes to market authorisation rather than a fundamental reform of the process. Far too many drugs and devices have sneaked onto the market with lower standards of scrutiny and the FDA's post-marketing surveillance isn't up to the task.
The FDA knew about rofecoxib's cardiovascular risks for years and did nothing; they've known about rosiglitazone for years and they've done very little. It's just too easy for bad products to get on the market and stay on the market under the FDA's less-than-vigilant gaze.
I think that's a matter of opinion, rather than a purely factual statement. It intersects with your comment "Far too many drugs and devices have sneaked onto the market with lower standards of scrutiny", because it begs the question "how many is too many?".
It's all a trade-off. The higher level of scrutiny requires more time and more money and can result in people being denied drugs that could have prevented harm or death.
It's my own personal opinion, but I think the FDA does a pretty good job considering the judgements they are asked to make.
And rofecoxib was eventually pulled, but only after careful scrutiny of the risk. It's not hard to do a retrospective analysis and see a safety signal. Pulling a product based on limited data means you may be doing harm by denying people a drug that helps them. Before the COX2s existed, perforated ulcers due to NSAID use weren't exactly rare.
And I'll just end with one final comment - overseeing a safe and effective drug supply is really damn hard. So I try and take that into consideration when evaluating the FDA's (or any countries drug agency) actions.
And it might be good to know that most countries approve drugs based on the same data - global clinical trials. Australia is looking at the same data as the US when making an approval decision. Some countries require local trials (China and India, I believe).
In addition, there is plenty of collaboration across drug approval bodies when developing or changing regulations. They are all talking to each other to determine the best approach.
Here is a good example of the FDA and TGA (Australia) and Canada approving a new cancer drug together[1]
[1] https://www.fdanews.com/articles/192766-fda-approves-first-c...