Depends on who's defining "nicely". If you live within its constraints, you can get a lot done. But one major difference is how inline math is formatted, especially with single-spaced body text. TeX derivatives tend to handle that sort of thing much better.

Example: Word can't squeeze a Fourier series (a sum and a set of fractions) into a single baseline of text, while simpler equations are no problem. Word places some (in my opinion) odd constraints on the height of the summation symbol, apparently forcing it to be as tall as the summed items, including parentheses and other grouping symbols. (EDIT: Philippe Goutet's Word experience is broader than mine -- you can get a more TeX-like inline summation, but for whatever reason, that's not the default in the Fourier series inline template.)

Word single-spaced:

LaTeX single-spaced:

Word double-spaced:

Word's typesetting algorithm for math is currently better than is TeX's (the situation might change with LuaTeX); the main improvement is in the spacing (factorials are automatically spaced correctly, for example).

But that's only part of the picture as, when using Word, you are still stuck with (at least) three main flaws:

1. Word handles the text badly. You say you don't care how efficient is TeX justification algorithm, but you should (especially if you want almost book level quality: Word is totally inadequate for that). Formulas are only a small part of the math document, so how the text is handled is very important. And I'm not even mentioning things like table of contents, bibliographies, index creation, automated numbering of theorems, etc. which are a real pain with Word and a great strength of LaTeX.

2. Word handles the interactions between text and formulas badly. For example, obtaining a numbered formula in Word 2007 (I've not tried Word's latest version, so I don't know if it has improved) is a real pain (see Word's blog "Equation Numbering" post ^[1]). And that's just one example. Remember that TeX/LaTeX was conceived to be used with maths, whereas for Word, it was just an afterthought.

3. If you want to do something complicated or which was not planned by Word's creators, you will be struck. For example, try to typeset a summation symbol with both a prime and sub/superscript. With LaTeX, you're sure it can be done and it's easy to find a package to do just that (googling "latex primed sum" gives the answer immediately).

LaTeX has a learning curve, of course, but you can get started in just one afternoon, and if you need help, it's easy to find. I made the transition from Word to LaTeX a few years ago and never regretted it. If you plan on writing one day a long report with maths (e.g. a PhD thesis), then choosing LaTeX is a no-brainer, and the sooner, the better for you.

The math typesetting in Word (post 2007) is closely modeled on TeX. See for example this quote ^[1] from the person who lead the team responsible for the Math typesetting in Word:

The TeXbook is a user manual that includes a detailed specification for mathematical typography. We have used many of its choices and methodology in creating our solutions, which are appropriately enhanced with the use of OpenType tables and some additional constructs. Although the TeX source code is available, it cannot be used directly for several reasons. First the code is like a web rather than being hierarchical and uses many global variables. This makes it cumbersome to employ in the instance-oriented contexts used at Microsoft. Complicating this is that TeX is a complete document imaging system, not one limited to mathematics. As such many aspects of the program that are used for mathematics are used also for other kinds of layout like headers, footers, figures, and footnotes. Extricating the mathematical algorithms from this web of code would be significantly harder than recreating the desired display quality using our own methodologies and the specifications given in The TeXbook, especially in Appendix G. Furthermore we want to take advantage of our OpenType math fonts to obtain better positioning of subscripts, superscripts, and other symbols than possible by default using TeX. Another complication is that Office is an international environment and our math facility needs to be compatible with all languages that we support, potentially simultaneously. We also want to take advantage of contemporary methods to optimize screen displays.

Conversely, since that time, the OpenType Math tables that Murray mentions were added [to Cambria Math] to support the typesetting of Math in Office have been added to other free math fonts and are read by TeX systems such as XeTeX with OpenType font support, so technological advances move both ways.

Personally I find a WYSIWYG word processing system a scary and hostile environment for editing documents, but apparently that feeling is far from universally shared, so in the end it comes down to personal preferences, which are hard to measure objectively.

I had a very traumatic experience with Microsoft Word + mathmatics in the past. :( That said, I think newer versions of Word are getting more intuitive and it seems the equation editor has now support for a subset of LaTeX commands.

Maybe you could investigate what's the general rule for typesetting documents in your university department. Since you will probably have to share your documents with colleagues and advisors, it's very important to discover your environment. I know professors that are quite feral with editors - bring a document format other than doc/docx and you are doomed. Yes, they exist in academia.

Sadly, I don't see any strong TeX tradition in Brazilian universities (thankfully a few mathematics and computer science departments are still in la résistance, but even there it's only a few people that use TeX). That might be a barrier. Take a look if your deparment has any TeX templates. A friendly warning: our crazy ABNT format is a pain, specially for TeX.

Anyway, I should advocate for TeX too, so here's a blog post I wrote in our community blog: More than typesetting ^[1]. Hope it helps. :)

To be honest: If you want to do maths, to some more extent than making your taxes right, you should learn LaTeX. Sooner or later you'll find that it's unavoidable.

You can start with the LyX system ^[1] that can help you with your documents and it can provide sufficient result, surely much better than Word does, even Word 2010, as Mike Renfro points out in his answer. LyX is based on LaTeX and allows many its features, while it is mostly a WYSIWYG editor. It is a good bridge from "pure WYSIWYG" like Word to "pure larguage" like LaTeX.

You will see that it is not that difficult and for basic documents (like scientific articles, your thesis or an extended homework) you don't need anything special generally, neither in LyX or in LaTeX itself.

One advantage of using LaTeX is: we cooperate a lot. If you get stuck, go and ask. The method of a minimal working example will most probable guide you to a solution. There is no MWE with word.

Well a simple reason for using (La)TeX over Word is not simply restricted to mathematics display. A lot of faffing about (i.e. time consuming minor changes) is often required in Word to keep the "look" right so you are not spending time on the content. Look how badly word handles placement of figures and their associated captions.

I use both Word (for interaction with my colleagues) but usually typeset my reports (for industry) using PDFLaTeX. The reason for using LaTeX for the reports is that I get a direct PDF output that the end user can't change. (I know I can print to a pdf printer from Word but sometimes the characters don't print as good/consistent...)

That's my tuppence worth...