Relation between Borchers class and the LSZ formula on S-matrix equivalence It seems well known that different quantum fields can give rise to the same $S$-matrix. I know of two ways this is described.
The first is through the Borchers class of relatively local fields, i.e. fields $A(x)$ and $B(x)$ which satisfy 
$$[A(x),B(y)] = 0,\ \ \ \ \text{$x$ and $y$ space-like separated.}$$
It is then known that all fields within the same Borchers class have the same $S$-matrix.
The second is through the LSZ reduction formula, where the only requirement on the field is that:


*

*It has vanishing vacuum expectation value $\langle 0|\phi(x)|0\rangle = 0$.

*It has non-zero overlap with the relevant single particle states $\langle p|\phi(x)|0\rangle \neq 0$.
And any field $\phi(x)$ which satisfies the two above conditions (up to normalization) will give the same $S$-matrix.
My question is a bit vague, but I would like a better understanding as to how the two criteria are related. The two doesn't seem equivalent. 
For example, let's consider the simple example of a scalar field $\phi(x)$. We can consider the normal ordered field $:\phi^2(x):$ which lies within the same Borchers class as $\phi(x)$. So we expect the two to give the same $S$-matrix. On the other hand, the overlap of $:\phi^2(x):$ with single particle states is vanishing, so we should expect a different $S$-matrix due to the LSZ formula.
In general, it doesn't seem true that 
$$\text{"relatively local" $\implies$ "non-zero single particle overlap",}$$ 
nor does it seem true that 
$$\text{"non-zero single particle overlap" $\implies$ "relatively local".}$$ 
So how do I reconcile these two (apparently inconsistent) ways of looking at $S$-matrix equivalence?
 A: The comment in my book (cited by Diffycue) originated in my own puzzlement   when I was a student and reading   my own favourite QFT book: that  by K. Nishijima, Fields and Particles: Field Theory and Dispersion Relations. In it he seems to use the LSZ technique show that any two fields $\phi_i$ and $\phi_2$ that have matrix elemements between the vacuum and the one particle states of the form
$$
\langle k\vert \phi_i(x)\vert 0\rangle = \sqrt Z_i e^{ikx}
$$
will have the same $S$-matrix. Since this is a property possesed by  $\phi_{\rm in}(x)$ and $\phi_{\rm out}(x)$, which are free, he seemed to have proved that all theories  were free. It took me ages and a very careful reading of his argument to realize that he  made use   of  Lorentz transformation properties of time-ordered correlators such as
$$
\langle 0\vert T\{ \phi_1(x_1) \phi_2(x_2)\ldots \}\vert0\rangle
$$ 
that  are messed up unless $\phi_1$ and $\phi_2$ commute outside the lightcone. Thus Nishijima was tacitly assuming the mutual locality of the various fields.  
As a consequence $\phi$ and $\phi+\phi^3$ will have the same $S$ matrix for $\lambda \phi^4$ theories with their ${\mathbb Z}_2$ symmetry, but $\phi$ and $\phi^2$ will not because  $\phi^2$ cannot couple the vacuum to the $\phi$ state because it has the wrong  ${\mathbb Z}_2$ parity. 
A: You seem to have a very good understanding of the LSZ formula; perhaps I am the one who is mistaken here. But I was under the impression that two fields in the same Borchers class were


*

*Mutually local

*Connect the vacuum to the same states.


I.e., the statement "the overlap between $: \phi^2 :$ and single-particle states vanishes" would contradict the statement that "$: \phi^2 : $ lies within the same Borchers class as $\phi$."
To me it seems like you are completely correct that locality does not imply nonzero overlap and vice versa, which is why we require condition 2. above.
I'll note also that in my favorite QFT book, The Physics of Quantum Fields by Michael Stone, that the author mentions: "The set of fields which [sic] connect the vacuum to the required states and are mutually local in that they commute at space-like separation is called a Borcher's class." But in the notes my QFT class used, there was no explicit mention of the connectivity property.
