You will not able to say “Yeah but I only smoke two cigarettes a day you know.”

There seems to be a plethora of studies showing that light smokers (even smoking less than 4 cigarettes a day) still face significantly higher risk of diseases than non-smokers. The article “Health Effects of Light and Intermittent Smoking” (Circulation, April 6, 2010 vol. 121 no. 13) reviews the many studies in the field, and the picture is quite damning.

I reproduce the table here, but you can access the full article freely. The important column is Risk: RR means relative risk, which roughly translate into how more likely is a light-smoker to develop the disease compared to a non-smoker. If the number is 1, it means light smokers and non-smokers face the same risks. The further away from 1 the number is, the shittier it is for light smokers.

Disease Level of Smoking Risk for Light Smokers vs Nonsmokers Study Design
Cig indicates cigarettes; RR, relative risk; OR, odds ratio; and RH, relative hazard.
Cardiovascular disease
    Ischemic heart disease28 1–4 cig/d RR 2.74 (2.07–3.61) in men; RR 2.94 (1.75–4.95) in women Prospective cohort
    Aortic aneurysm30 <10 cig/d RR 2.29 Prospective cohort
    Cardiovascular mortality31 Occasional smoking RR 1.5 in men (1.0–2.3) Prospective cohort
Malignancy
    Esophageal cancer32 1–14 cig/d RR 4.25 Prospective cohort
    Lung cancer28 1–4 cig/d RR 2.79 (0.94–8.28) in men; RR 5.03 (1.81 to 13.98) in women Prospective cohort
    Gastric cancer33 1–4 cig/d RR 2.4 (1.3–4.3) Case control
    Pancreatic cancer34 <10 cig/d RR 1.8 (1.4–2.5) Prospective cohort
Respiratory diseases
    Lower respiratory tract infections35 Light smoking (<1 pack/d) RR:1.5 in men; RR 1.13 in women Prospective cohort
    Prolonged duration of respiratory symptoms: cough35 <1 pack/d Duration of respiratory symptoms (cough) was 7.7 d in the light smoking group vs 6.8 d in never smokers Prospective cohort
Reproductive health
    Impaired fecundity in women (delayed time to conception)36 1–4 cig/d Increasing OR of delayed conception from 1.1 for 6-mo delay to 3.2 at 18-mo delay Prospective cohort
    Spermatozoa function36 4 cig/d for 5 y Spermatozoa showed decreased density/motility Prospective cohort
    Placenta previa37 Light smoking (<1 pack/d) OR 2.2 (0.87–7.83) Case control
    Ectopic pregnancy38 <10 cig/day OR 1.4 (0.8–2.5) Case control
Other conditions
    Ankle fractures in women39 1–10 cig/d OR 3.0 (1.9–4.6) Retrospective
    Cataracts/development of nuclear lens opacities40 Light smoking (<10 cig/d) OR 1.68 (1.14–2.49) Prospective cohort
Physical disability after meniscal tear41 Light smoking (<1 pack/d) RH 1.44 (1.07–1.94) Prospective cohort
All-cause mortality
    Risk of all-cause mortality in men31 Occasional smoking OR 1.6 (1.3–2.1) Prospective cohort

The table offers a synthesis of many studies found in the scientific literature addressing the issue of light smokers. There are limitations as to what those studies can do, for example one cannot devise a randomised controlled trial of smoking for obvious ethical reasons. But suffice to say that most of the studies they cite do control for confounding factors such as diet, exercise, wealth, etc.