Placing the mics next to one another doesn't give exactly the same results as vertically aligned mics but with small diaphragm mics which may be placed with their diaphragms physically very close together, in all but the most critical applications, it's all but impossible to tell between vertically aligned and side by side setups.
In larger diaphragm/large bodied mics like C414s, U89s, etc., where the larger bodies and capsule housings make it harder to get the capsules close together (and where it's harder to mount them vertically without resorting to modified/specially built hardware) you may get more problems. In practice, many large diaphragm mic designs actually position the capsule closer to the top of the housing than the bottom - which moves it slightly away from the effects of body, more into free space, and makes it easier to achieve close capsule mounting to a similar mic when vertically aligned.
The main differences between side by side and vertical mounting are in the nature of the inter-channel delays and reflections/shadowing caused by the mic bodies and mounts.
Ideally, for Blumlein, to minimise interchannel delays, the capsules should be vertically mounted, however, with physically small mics mounted side by side, the interchannel delays may actually be smaller than with some larger mics, mounted vertically. The only difference then is that, with vertically mounted mics, the interchannel delays are greatest in the vertical alignment so (in theory) impact less on the accuracy/stability of the horizontal stereo image (on the basis that people are less sensitive to variations in median plane auditory imaging) . Side by side mounting increases the horizontal interchannel delays so impacts more on the percieved stereo image.
As for shadowing/reflection, if figure-8 mic capsules are mounted vertically above one another, the closest pieces of hardware are in the response null of the polar pattern, and the bodies are physically as far apart as possible, so reflection effects are minimised. Mounting fig-8 mics side by side in anything other than a parallel configuration (e.g. a
Faulkner Array ), causes one mic to some extent to shadow the polar pattern of the other and can result in reflections off the structure of one mic interfering with sound entering the other. Improvised, or less than optimally designed mouting systems can also cause reflection problems but whether or not, in practice, these are audible/problematic is another matter.
Purpose built coincident stereo mics like the C422/C426, SM69, VM1S, MC742, etc., etc., have an advantage (apart from practicality) over using pairs of the mono equivalent mics in that they reduce the amount of hardware around the capsules - one mic body, no need for complex stereo mounts, etc..
The strength of these various delay/shadowing/reflection effects varies from setup to setup and, at it's best, some side by side mounts will be less problematice than some of the worst affected vertically aligned pairs but, in general, vertical mounting of a given pair of mics will always achieve fewer errors/more perfect results in Blumlein/MS than side by side mounting.
Having said all that, the differences between vertical and side by side mounting normally are small and practicalities like availability of mics and mounting hardware, together with positioning and rigging considerations usually outweigh any advantages/disadvantages inherent in the chosen mounting method. In other words - side by side isn't the best way to do Blumlein or MS but, done with care and a little consideration of the details, doesn't usually make enough difference to bother anyone so do whatever works for you. It might, however, be technically wrong, to describe anything other than vertically mounted fig-8s as 'Blumlein', in the same way that patterns other than cardioid can't be used to produce true 'ORTF' arrays. (It might not be wrong as Blumlein specifies co-incidence, which is physically impossible, and side by side may, in distance, be as (or more) physically 'coincident' as vertical mounting. I'm not sure whether Blumlein's description differentiates between horizontal and vertical mounting - I'll check when I have some time.)