This paper examines the accuracy of design equations specified in the North American, European and Australasian codes for cold-formed steel structures in determining the net section tension capacity of bolted connections in flat steel sheets. It points out that the shear lag factors embedded in the code equations either yield "anomalous" results or become irrelevant when they exceed unity. The "anomaly" was demonstrated through laboratory tests and is explained by using simple calculus. The configurations of specimens tested in the laboratory include single shear- and double-shear connections, with single or double bolts in a line parallel or perpendicular to the force. A proper mathematical expression for the in-plane shear lag factor, which does not suffer from the anomaly of the code equations and never implies shear lag factors greater than unity for any configuration, is presented and shown to yield improved results compared to the current specifications. The resistance factor of 0.8 for the proposed equation is computed with respect to the load and resistance factor design (LRFD) approach specified in the North American specification for the design of cold-formed steel structures.