Quantum physics applies to all events in the Universe, whether they're microscopic or not. However, for macroscopic events, classical physics (approximately) holds, too. In other words, one may show that whenever the relevant products of powers of quantities that describe a system are much greater than $\hbar$, Planck's constant, we may use approximate laws to describe the behavior of the system, namely the laws that are formally the $\hbar\to 0$ limit of the laws of quantum physics: the corresponding classical theory.
The emergence of classical physics as a limit of a quantum mechanical theory has many conceptual and mathematical aspects, see e.g.
http://motls.blogspot.com/2011/11/how-classical-fields-particles-emerge.html?m=1
Concerning your particular question on tunneling, macroscopic objects may tunnel through the wall; it's just very unlikely. Consider a pair of macroscopic objects, a wall and a ball. Each of them has $10^{25}$ particles. If the microscopic probability of an atom's penetration through another atom is $q$, then the penetration of the two macroscopic objects through each other is pretty much the "penetration of each atom of a ball through each atom of a wall" and the probability is therefore
$${\Large P\sim q^{10^{25}\times 10^{25}} = q^{10^{50}} \sim \exp(-10^{50})} $$
if $q\sim \exp(-1)$ is a reasonable suppression for the microscopic process. Let me emphasize that the probability $P$ above has the form $0.000\cdots 0001$ and the number of zeroes after the decimal point in this sequence of digits isn't just fifty. Instead, the number of zeroes in the tiny number is $10^{50}$ i.e. 100,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000. It is a ludicrously tiny number which is why the tunneling of macroscopic bodies may be neglected and considered impossible. The largest object for which such a tunneling could be meaningfully considered would be a biological molecule such as a protein.
