It would probably be much easier to think about this question if you clearly defined what you mean by effective nuclear charge.
According to my understanding, the effective nuclear charge $Z_\mathrm{eff}$ is a fitting parameter appearing when you try to model your valence electrons by a hydrogen orbital.
In this sense, there's two effects in play: $Z_\mathrm{eff}$ will increase as the total charge of the nucleus increases, but $Z_\mathrm{eff}$ will decrease from the screening effects of the non-valence electrons (there's also some weak screening due to other valence electrons but that's a much smaller effect).
If you consider let's say the valence electrons of Na (well, the single one of them), then it will feel the positive charge of $+11$ from the nucleus, shielded by the negative charge $-10$ of the core electrons, leading to an effective nuclear charge somewhat larger than $+1$ because the screening is not perfect (in fact, $Z_\mathrm{eff}$ is about 2 for the valence shell of Na).
Then, consider Cl in the same row. The valence electrons will feel $+17$ charge from the nucleus, shielded by $-10$ valence electrons, meaning $Z_\mathrm{eff}$ will be in the order of magnitude of 7.
So now that we have established how to think about it, we can easily answer your question.
Compare Na to K. Both of these atoms will have one single valence electrons, so the effective nuclear charge is about $+1$, with the differences accounting for the efficiency of screening.
In general, the closer to the core an electron is, the more effectively it screens the nucleus. Meaning that a $2p$ electron will contribute more to screening than a $3p$ electron. This really all comes down to the radius of the orbital.
The core electrons in K are more diffuse than the ones in Na, so they don't screen as efficiently.
Therefore $Z_\mathrm{eff}$ is larger for the valence electron of K than for the valence electron of Na $\rightarrow$ $Z_\mathrm{eff}$ increases as you go down a group.