Made in the US: in the distant past I found the Acoustics Resdearch turtables free from rumble , wow & flutter. They need the occasional oil drop at the base of the spindle plus occasionally a new dfive belt much like their British sucessor, the Linn Sondek . The weakest link in the sound chains as always will be the electro-mechanical devices (pickup & speakers (or headphones). were US companies making decent stuff, but at present I assume most is imported from cheaper sources-
The Danish company Ortofon still makes decent "needle microphones" but like all pickups will need a preamp to boost the signal and a RIIA filter to compensate for the tricks needed to create the record. so that it is ready to be listened to. Any TT with a line-signal out must have this inside. The ones with USB out have an additional analog to digital converter which may or may not be the stropongest or weakest link in the chain onwards.
If you can handle a saw, glue and woodscrews chances are that you can build a decent closed box Loudspeaker large enough not to require much power from your amplifier and still give decent bass without the box resonating (I must admit that when I built my bass cabinets, I used a double-wall, sand filled construction to avoid resonances.
Tricks like having small diameter, long voicecoil, speakers with high mass & compensating the 6dB/octave bass rolloff with filters will require more power and may create strange distortions. Easily driven bass reflex (boxes with a hole) that superimpose the free air resonance of the speaker on the resonance of speaker mounted in the box+ hole need less power, but much more care in calculating, damping and measuring, not to speak of the parasitic tones (added flexing) of lighweight and soft speaker cones which makes the choice of speaker elements a difficult task while internal "crossover " filters to distribute current to voice coils according to the frequency the element handles best can be another point where detrimental savings are made.
All something which need a bit of time and scepticism in the face of glib ads with zero scientific backing-. A particularly bad example is set by a claim of say 1oow from a power amp but not divulging whether it is measured on a random full audible spectrum signal or just on one frequency, and whether it is a continuous signal (watts RMS) or a short pulse-
p.